diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..7d4a1bd7d --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +c1ffff8904452cfbdcfa2438656350764e533fe4ad59f7b8859706fb27985867 \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_tensor_constraints.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py new file mode 100644 index 000000000..5502ab151 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py @@ -0,0 +1,2725 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + input_5 = torch._C._nn.linear( + x_412, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_, + ) + x_412 = ( + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = None + input_6 = torch._C._nn.gelu(input_5) + input_5 = None + x_413 = torch.nn.functional.dropout(input_6, 0.0, False, False) + input_6 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py new file mode 100644 index 000000000..207e9350e --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py @@ -0,0 +1,3490 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_" + shape = [1536, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt b/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt new file mode 100644 index 000000000..0058e5ef8 --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt @@ -0,0 +1 @@ +4b84b13cf6edfc2284f331776426f1742970193e1d202ffc898a19f4ec8fc5a9 \ No newline at end of file diff --git a/samples/timm/convnext_nano.d1h_in1k/graph_net.json b/samples/timm/convnext_nano.d1h_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_nano.d1h_in1k/input_meta.py b/samples/timm/convnext_nano.d1h_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_nano.d1h_in1k/input_tensor_constraints.py b/samples/timm/convnext_nano.d1h_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_nano.d1h_in1k/model.py b/samples/timm/convnext_nano.d1h_in1k/model.py new file mode 100644 index 000000000..33e949b77 --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/model.py @@ -0,0 +1,1279 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + reshape_12 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_153 = x_152.mul(reshape_12) + x_152 = reshape_12 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + reshape_13 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_164 = x_163.mul(reshape_13) + x_163 = reshape_13 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnext_nano.d1h_in1k/weight_meta.py b/samples/timm/convnext_nano.d1h_in1k/weight_meta.py new file mode 100644 index 000000000..3985ec06f --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/weight_meta.py @@ -0,0 +1,1490 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt b/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt new file mode 100644 index 000000000..5d1479c7d --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt @@ -0,0 +1 @@ +4b04e033486ab9dcce14edbf03a9da60666faabdba55185f29cb9b0e5e8781d5 \ No newline at end of file diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json b/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/input_meta.py b/samples/timm/convnext_nano_ols.d1h_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/input_tensor_constraints.py b/samples/timm/convnext_nano_ols.d1h_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/model.py b/samples/timm/convnext_nano_ols.d1h_in1k/model.py new file mode 100644 index 000000000..2a035e4cc --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_5 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + reshape_12 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_153 = x_152.mul(reshape_12) + x_152 = reshape_12 = None + x_154 = x_153 + input_5 + x_153 = input_5 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + reshape_13 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_164 = x_163.mul(reshape_13) + x_163 = reshape_13 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py b/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py new file mode 100644 index 000000000..67e013662 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py @@ -0,0 +1,1510 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_pico.d1_in1k/graph_hash.txt b/samples/timm/convnext_pico.d1_in1k/graph_hash.txt new file mode 100644 index 000000000..3a8ac56d9 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +d4f4098bb40e1e23a759cde6faf233d3399e1878b3baa0fd14097af0c6316598 \ No newline at end of file diff --git a/samples/timm/convnext_pico.d1_in1k/graph_net.json b/samples/timm/convnext_pico.d1_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_pico.d1_in1k/input_meta.py b/samples/timm/convnext_pico.d1_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_pico.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_pico.d1_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_pico.d1_in1k/model.py b/samples/timm/convnext_pico.d1_in1k/model.py new file mode 100644 index 000000000..11dd6cdd9 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/model.py @@ -0,0 +1,1123 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_pico.d1_in1k/weight_meta.py b/samples/timm/convnext_pico.d1_in1k/weight_meta.py new file mode 100644 index 000000000..5449045a4 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/weight_meta.py @@ -0,0 +1,1310 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt b/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt new file mode 100644 index 000000000..a3afbfa7c --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +2c0f4ae67241056d833f6493124e5a0295849c51c5c9cf26e5a4d988a38b6b3a \ No newline at end of file diff --git a/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json b/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_pico_ols.d1_in1k/input_meta.py b/samples/timm/convnext_pico_ols.d1_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_pico_ols.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_pico_ols.d1_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_pico_ols.d1_in1k/model.py b/samples/timm/convnext_pico_ols.d1_in1k/model.py new file mode 100644 index 000000000..47c87c9f2 --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/model.py @@ -0,0 +1,1143 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_5 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_5 + x_131 = input_5 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py b/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py new file mode 100644 index 000000000..3dcb9f67d --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py @@ -0,0 +1,1330 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_small.fb_in1k/graph_hash.txt b/samples/timm/convnext_small.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..1ecc6e186 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +0b86a61c951a12e51f0f7e01e08306b14026c01e6a6f636521e61c790236501f \ No newline at end of file diff --git a/samples/timm/convnext_small.fb_in1k/graph_net.json b/samples/timm/convnext_small.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_small.fb_in1k/input_meta.py b/samples/timm/convnext_small.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_small.fb_in1k/input_tensor_constraints.py b/samples/timm/convnext_small.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_small.fb_in1k/model.py b/samples/timm/convnext_small.fb_in1k/model.py new file mode 100644 index 000000000..4d3387279 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_small.fb_in1k/weight_meta.py b/samples/timm/convnext_small.fb_in1k/weight_meta.py new file mode 100644 index 000000000..4444eb560 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt b/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..b81f0c333 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +49bda9b465818c7f5416d39636d24cdae7ea5d056153a366e1f60fe453b56e6e \ No newline at end of file diff --git a/samples/timm/convnext_tiny.fb_in1k/graph_net.json b/samples/timm/convnext_tiny.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_tiny.fb_in1k/input_meta.py b/samples/timm/convnext_tiny.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_tiny.fb_in1k/input_tensor_constraints.py b/samples/timm/convnext_tiny.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_tiny.fb_in1k/model.py b/samples/timm/convnext_tiny.fb_in1k/model.py new file mode 100644 index 000000000..1ff70532d --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/model.py @@ -0,0 +1,1447 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = torch._C._nn.linear( + x_182, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_182 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = x_184.permute(0, 3, 1, 2) + x_184 = None + reshape_15 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_186 = x_185.mul(reshape_15) + x_185 = reshape_15 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = x_195.permute(0, 3, 1, 2) + x_195 = None + reshape_16 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_197 = x_196.mul(reshape_16) + x_196 = reshape_16 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_204 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + x_207 = x_206.permute(0, 3, 1, 2) + x_206 = None + reshape_17 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_208 = x_207.mul(reshape_17) + x_207 = reshape_17 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnext_tiny.fb_in1k/weight_meta.py b/samples/timm/convnext_tiny.fb_in1k/weight_meta.py new file mode 100644 index 000000000..075d12cb4 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/weight_meta.py @@ -0,0 +1,1850 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt new file mode 100644 index 000000000..e13286982 --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt @@ -0,0 +1 @@ +0d3941534699169e11117d7d410fb329687499ec49c1026f9adc7c4f15cbaa83 \ No newline at end of file diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/input_meta.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/input_tensor_constraints.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py new file mode 100644 index 000000000..9d9250ee8 --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py @@ -0,0 +1,1591 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_pre_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_pre_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_norm_pre_parameters_weight_ = ( + L_self_modules_norm_pre_parameters_weight_ + ) + l_self_modules_norm_pre_parameters_bias_ = ( + L_self_modules_norm_pre_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = x_27.permute(0, 3, 1, 2) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_30 = torch._C._nn.gelu(x_29) + x_29 = None + x_31 = torch.nn.functional.dropout(x_30, 0.0, False, False) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = x_52.permute(0, 3, 1, 2) + x_52 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_55 = torch._C._nn.gelu(x_54) + x_54 = None + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = x_143.permute(0, 3, 1, 2) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145) + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = x_154.permute(0, 3, 1, 2) + x_154 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156) + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = x_165.permute(0, 3, 1, 2) + x_165 = None + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167) + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_182 = torch._C._nn.gelu(x_181) + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + reshape_15 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_186 = x_185.mul(reshape_15) + x_185 = reshape_15 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = x_190.permute(0, 3, 1, 2) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_193 = torch._C._nn.gelu(x_192) + x_192 = None + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = torch.conv2d( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + reshape_16 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_197 = x_196.mul(reshape_16) + x_196 = reshape_16 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = x_201.permute(0, 3, 1, 2) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_204 = torch._C._nn.gelu(x_203) + x_203 = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + reshape_17 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_208 = x_207.mul(reshape_17) + x_207 = reshape_17 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = x_209.permute(0, 2, 3, 1) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (768,), + l_self_modules_norm_pre_parameters_weight_, + l_self_modules_norm_pre_parameters_bias_, + 1e-06, + ) + x_210 = ( + l_self_modules_norm_pre_parameters_weight_ + ) = l_self_modules_norm_pre_parameters_bias_ = None + x_212 = x_211.permute(0, 3, 1, 2) + x_211 = None + x_213 = torch.nn.functional.adaptive_avg_pool2d(x_212, 1) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py new file mode 100644 index 000000000..c5a0ff17d --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py @@ -0,0 +1,1850 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_pre_parameters_weight_: + name = "L_self_modules_norm_pre_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_pre_parameters_bias_: + name = "L_self_modules_norm_pre_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt b/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt new file mode 100644 index 000000000..3a8a8f315 --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt @@ -0,0 +1 @@ +e09f137b683383f74a7c3ec87bf480a8068e0212c4950307376b464a21869dcb \ No newline at end of file diff --git a/samples/timm/convnext_xlarge.fb_in22k/graph_net.json b/samples/timm/convnext_xlarge.fb_in22k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_xlarge.fb_in22k/input_meta.py b/samples/timm/convnext_xlarge.fb_in22k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_xlarge.fb_in22k/input_tensor_constraints.py b/samples/timm/convnext_xlarge.fb_in22k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_xlarge.fb_in22k/model.py b/samples/timm/convnext_xlarge.fb_in22k/model.py new file mode 100644 index 000000000..b56881e8d --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (256,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (256,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (512,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1024,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2048,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py b/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py new file mode 100644 index 000000000..5b3472b41 --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [256, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2048, 1024, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [21841, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [21841] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt new file mode 100644 index 000000000..6891d7f77 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt @@ -0,0 +1 @@ +c484ac258f449ea7468ea4628d2c98a514c2dec856c9b004245122a56539afe2 \ No newline at end of file diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_meta.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_tensor_constraints.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py new file mode 100644 index 000000000..893cd41e4 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py @@ -0,0 +1,2987 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (384,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + x_74 = torch.nn.functional.layer_norm( + x_73, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_74 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_76 = torch._C._nn.gelu(x_75) + x_75 = None + x_77 = torch.nn.functional.dropout(x_76, 0.0, False, False) + x_76 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_79.permute(0, 3, 1, 2) + x_79 = None + reshape_6 = l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_81 = x_80.mul(reshape_6) + x_80 = reshape_6 = None + x_82 = x_81 + x_71 + x_81 = x_71 = None + x_83 = x_82.permute(0, 2, 3, 1) + x_82 = None + x_84 = torch.nn.functional.layer_norm( + x_83, + (768,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_85 = x_84.permute(0, 3, 1, 2) + x_84 = None + input_3 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_86 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + input_3 + x_95 = input_3 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-05, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-05, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-05, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-05, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-05, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-05, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-05, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-05, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-05, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-05, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-05, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-05, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-05, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = torch.conv2d( + x_371, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_373 = x_372.permute(0, 2, 3, 1) + x_372 = None + x_374 = torch.nn.functional.layer_norm( + x_373, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-05, + ) + x_373 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_376 = torch._C._nn.gelu(x_375) + x_375 = None + x_377 = torch.nn.functional.dropout(x_376, 0.0, False, False) + x_376 = None + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_379 = torch.nn.functional.dropout(x_378, 0.0, False, False) + x_378 = None + x_380 = x_379.permute(0, 3, 1, 2) + x_379 = None + reshape_33 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_381 = x_380.mul(reshape_33) + x_380 = reshape_33 = None + x_382 = x_381 + x_371 + x_381 = x_371 = None + x_383 = torch.conv2d( + x_382, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ = (None) + x_384 = x_383.permute(0, 2, 3, 1) + x_383 = None + x_385 = torch.nn.functional.layer_norm( + x_384, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_, + 1e-05, + ) + x_384 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ = (None) + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_387 = torch._C._nn.gelu(x_386) + x_386 = None + x_388 = torch.nn.functional.dropout(x_387, 0.0, False, False) + x_387 = None + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_390 = torch.nn.functional.dropout(x_389, 0.0, False, False) + x_389 = None + x_391 = x_390.permute(0, 3, 1, 2) + x_390 = None + reshape_34 = l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ = ( + None + ) + x_392 = x_391.mul(reshape_34) + x_391 = reshape_34 = None + x_393 = x_392 + x_382 + x_392 = x_382 = None + x_394 = torch.conv2d( + x_393, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ = (None) + x_395 = x_394.permute(0, 2, 3, 1) + x_394 = None + x_396 = torch.nn.functional.layer_norm( + x_395, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_, + 1e-05, + ) + x_395 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ = (None) + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_398 = torch._C._nn.gelu(x_397) + x_397 = None + x_399 = torch.nn.functional.dropout(x_398, 0.0, False, False) + x_398 = None + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_401 = torch.nn.functional.dropout(x_400, 0.0, False, False) + x_400 = None + x_402 = x_401.permute(0, 3, 1, 2) + x_401 = None + reshape_35 = l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ = ( + None + ) + x_403 = x_402.mul(reshape_35) + x_402 = reshape_35 = None + x_404 = x_403 + x_393 + x_403 = x_393 = None + x_405 = torch.conv2d( + x_404, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ = (None) + x_406 = x_405.permute(0, 2, 3, 1) + x_405 = None + x_407 = torch.nn.functional.layer_norm( + x_406, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_, + 1e-05, + ) + x_406 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ = (None) + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_409 = torch._C._nn.gelu(x_408) + x_408 = None + x_410 = torch.nn.functional.dropout(x_409, 0.0, False, False) + x_409 = None + x_411 = torch._C._nn.linear( + x_410, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_412 = torch.nn.functional.dropout(x_411, 0.0, False, False) + x_411 = None + x_413 = x_412.permute(0, 3, 1, 2) + x_412 = None + reshape_36 = l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ = ( + None + ) + x_414 = x_413.mul(reshape_36) + x_413 = reshape_36 = None + x_415 = x_414 + x_404 + x_414 = x_404 = None + x_416 = x_415.permute(0, 2, 3, 1) + x_415 = None + x_417 = torch.nn.functional.layer_norm( + x_416, + (1536,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_416 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_418 = x_417.permute(0, 3, 1, 2) + x_417 = None + input_4 = torch.conv2d( + x_418, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_418 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_419 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_420 = x_419.permute(0, 2, 3, 1) + x_419 = None + x_421 = torch.nn.functional.layer_norm( + x_420, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_420 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_422 = torch._C._nn.linear( + x_421, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_421 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_423 = torch._C._nn.gelu(x_422) + x_422 = None + x_424 = torch.nn.functional.dropout(x_423, 0.0, False, False) + x_423 = None + x_425 = torch._C._nn.linear( + x_424, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_424 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_426 = torch.nn.functional.dropout(x_425, 0.0, False, False) + x_425 = None + x_427 = x_426.permute(0, 3, 1, 2) + x_426 = None + reshape_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_428 = x_427.mul(reshape_37) + x_427 = reshape_37 = None + x_429 = x_428 + input_4 + x_428 = input_4 = None + x_430 = torch.conv2d( + x_429, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_431 = x_430.permute(0, 2, 3, 1) + x_430 = None + x_432 = torch.nn.functional.layer_norm( + x_431, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_431 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_433 = torch._C._nn.linear( + x_432, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_432 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_434 = torch._C._nn.gelu(x_433) + x_433 = None + x_435 = torch.nn.functional.dropout(x_434, 0.0, False, False) + x_434 = None + x_436 = torch._C._nn.linear( + x_435, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_435 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_437 = torch.nn.functional.dropout(x_436, 0.0, False, False) + x_436 = None + x_438 = x_437.permute(0, 3, 1, 2) + x_437 = None + reshape_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_439 = x_438.mul(reshape_38) + x_438 = reshape_38 = None + x_440 = x_439 + x_429 + x_439 = x_429 = None + x_441 = torch.conv2d( + x_440, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_442 = x_441.permute(0, 2, 3, 1) + x_441 = None + x_443 = torch.nn.functional.layer_norm( + x_442, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_442 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_444 = torch._C._nn.linear( + x_443, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_443 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_445 = torch._C._nn.gelu(x_444) + x_444 = None + x_446 = torch.nn.functional.dropout(x_445, 0.0, False, False) + x_445 = None + x_447 = torch._C._nn.linear( + x_446, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_446 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_448 = torch.nn.functional.dropout(x_447, 0.0, False, False) + x_447 = None + x_449 = x_448.permute(0, 3, 1, 2) + x_448 = None + reshape_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_450 = x_449.mul(reshape_39) + x_449 = reshape_39 = None + x_451 = x_450 + x_440 + x_450 = x_440 = None + x_452 = torch.nn.functional.adaptive_avg_pool2d(x_451, 1) + x_451 = None + x_453 = x_452.permute(0, 2, 3, 1) + x_452 = None + x_454 = torch.nn.functional.layer_norm( + x_453, + (3072,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-05, + ) + x_453 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_455 = x_454.permute(0, 3, 1, 2) + x_454 = None + x_456 = x_455.flatten(1, -1) + x_455 = None + x_457 = torch.nn.functional.dropout(x_456, 0.0, False, False) + x_456 = None + x_458 = torch._C._nn.linear( + x_457, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_457 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_458,) diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py new file mode 100644 index 000000000..4caff7eff --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [384, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [3072, 1536, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1024, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt new file mode 100644 index 000000000..3074310a1 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt @@ -0,0 +1 @@ +314f8e9539d94cb83851ab0d0b0288d5a96b09df0a76049d52406bbce1da79e5 \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_meta.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_tensor_constraints.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py new file mode 100644 index 000000000..551296ee7 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py @@ -0,0 +1,983 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + v = torch.var(x, dim=(-1,), keepdim=True) + add = v + 1e-06 + v = None + rsqrt = torch.rsqrt(add) + add = None + x_1 = x * rsqrt + x = rsqrt = None + x_2 = x_1 * l_self_modules_stem_modules_1_parameters_weight_ + x_1 = l_self_modules_stem_modules_1_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + x_4 = torch.conv2d( + x_3, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_5 = x_4.permute(0, 2, 3, 1) + x_4 = None + v_1 = torch.var(x_5, dim=(-1,), keepdim=True) + add_1 = v_1 + 1e-06 + v_1 = None + rsqrt_1 = torch.rsqrt(add_1) + add_1 = None + x_6 = x_5 * rsqrt_1 + x_5 = rsqrt_1 = None + x_7 = ( + x_6 + * l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_8 = x_7.permute(0, 3, 1, 2) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9) + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_14 = x_13.mul(reshape) + x_13 = reshape = None + x_15 = x_14 + x_3 + x_14 = x_3 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_17 = x_16.permute(0, 2, 3, 1) + x_16 = None + v_2 = torch.var(x_17, dim=(-1,), keepdim=True) + add_3 = v_2 + 1e-06 + v_2 = None + rsqrt_2 = torch.rsqrt(add_3) + add_3 = None + x_18 = x_17 * rsqrt_2 + x_17 = rsqrt_2 = None + x_19 = ( + x_18 + * l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_20 = x_19.permute(0, 3, 1, 2) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_22 = torch._C._nn.gelu(x_21) + x_21 = None + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_26 = x_25.mul(reshape_1) + x_25 = reshape_1 = None + x_27 = x_26 + x_15 + x_26 = x_15 = None + x_28 = x_27.permute(0, 2, 3, 1) + x_27 = None + v_3 = torch.var(x_28, dim=(-1,), keepdim=True) + add_5 = v_3 + 1e-06 + v_3 = None + rsqrt_3 = torch.rsqrt(add_5) + add_5 = None + x_29 = x_28 * rsqrt_3 + x_28 = rsqrt_3 = None + x_30 = ( + x_29 + * l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + ) + x_29 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + input_2 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_32 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_33 = x_32.permute(0, 2, 3, 1) + x_32 = None + v_4 = torch.var(x_33, dim=(-1,), keepdim=True) + add_6 = v_4 + 1e-06 + v_4 = None + rsqrt_4 = torch.rsqrt(add_6) + add_6 = None + x_34 = x_33 * rsqrt_4 + x_33 = rsqrt_4 = None + x_35 = ( + x_34 + * l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_36 = x_35.permute(0, 3, 1, 2) + x_35 = None + x_37 = torch.conv2d( + x_36, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_36 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_38 = torch._C._nn.gelu(x_37) + x_37 = None + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_41 = torch.nn.functional.dropout(x_40, 0.0, False, False) + x_40 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_42 = x_41.mul(reshape_2) + x_41 = reshape_2 = None + x_43 = x_42 + input_2 + x_42 = input_2 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_45 = x_44.permute(0, 2, 3, 1) + x_44 = None + v_5 = torch.var(x_45, dim=(-1,), keepdim=True) + add_8 = v_5 + 1e-06 + v_5 = None + rsqrt_5 = torch.rsqrt(add_8) + add_8 = None + x_46 = x_45 * rsqrt_5 + x_45 = rsqrt_5 = None + x_47 = ( + x_46 + * l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_50 = torch._C._nn.gelu(x_49) + x_49 = None + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_54 = x_53.mul(reshape_3) + x_53 = reshape_3 = None + x_55 = x_54 + x_43 + x_54 = x_43 = None + x_56 = x_55.permute(0, 2, 3, 1) + x_55 = None + v_6 = torch.var(x_56, dim=(-1,), keepdim=True) + add_10 = v_6 + 1e-06 + v_6 = None + rsqrt_6 = torch.rsqrt(add_10) + add_10 = None + x_57 = x_56 * rsqrt_6 + x_56 = rsqrt_6 = None + x_58 = ( + x_57 + * l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + ) + x_57 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = (None) + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + input_3 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_60 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_61 = x_60.permute(0, 2, 3, 1) + x_60 = None + v_7 = torch.var(x_61, dim=(-1,), keepdim=True) + add_11 = v_7 + 1e-06 + v_7 = None + rsqrt_7 = torch.rsqrt(add_11) + add_11 = None + x_62 = x_61 * rsqrt_7 + x_61 = rsqrt_7 = None + x_63 = ( + x_62 + * l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_4) + x_69 = reshape_4 = None + x_71 = x_70 + input_3 + x_70 = input_3 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + v_8 = torch.var(x_73, dim=(-1,), keepdim=True) + add_13 = v_8 + 1e-06 + v_8 = None + rsqrt_8 = torch.rsqrt(add_13) + add_13 = None + x_74 = x_73 * rsqrt_8 + x_73 = rsqrt_8 = None + x_75 = ( + x_74 + * l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_76 = x_75.permute(0, 3, 1, 2) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_78 = torch._C._nn.gelu(x_77) + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_82 = x_81.mul(reshape_5) + x_81 = reshape_5 = None + x_83 = x_82 + x_71 + x_82 = x_71 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_85 = x_84.permute(0, 2, 3, 1) + x_84 = None + v_9 = torch.var(x_85, dim=(-1,), keepdim=True) + add_15 = v_9 + 1e-06 + v_9 = None + rsqrt_9 = torch.rsqrt(add_15) + add_15 = None + x_86 = x_85 * rsqrt_9 + x_85 = rsqrt_9 = None + x_87 = ( + x_86 + * l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = (None) + x_88 = x_87.permute(0, 3, 1, 2) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_94 = x_93.mul(reshape_6) + x_93 = reshape_6 = None + x_95 = x_94 + x_83 + x_94 = x_83 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_97 = x_96.permute(0, 2, 3, 1) + x_96 = None + v_10 = torch.var(x_97, dim=(-1,), keepdim=True) + add_17 = v_10 + 1e-06 + v_10 = None + rsqrt_10 = torch.rsqrt(add_17) + add_17 = None + x_98 = x_97 * rsqrt_10 + x_97 = rsqrt_10 = None + x_99 = ( + x_98 + * l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_7) + x_105 = reshape_7 = None + x_107 = x_106 + x_95 + x_106 = x_95 = None + x_108 = x_107.permute(0, 2, 3, 1) + x_107 = None + v_11 = torch.var(x_108, dim=(-1,), keepdim=True) + add_19 = v_11 + 1e-06 + v_11 = None + rsqrt_11 = torch.rsqrt(add_19) + add_19 = None + x_109 = x_108 * rsqrt_11 + x_108 = rsqrt_11 = None + x_110 = ( + x_109 + * l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + ) + x_109 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + input_4 = torch.conv2d( + x_111, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_112 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_113 = x_112.permute(0, 2, 3, 1) + x_112 = None + v_12 = torch.var(x_113, dim=(-1,), keepdim=True) + add_20 = v_12 + 1e-06 + v_12 = None + rsqrt_12 = torch.rsqrt(add_20) + add_20 = None + x_114 = x_113 * rsqrt_12 + x_113 = rsqrt_12 = None + x_115 = ( + x_114 + * l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_114 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_118 = torch._C._nn.gelu(x_117) + x_117 = None + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + reshape_8 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_122 = x_121.mul(reshape_8) + x_121 = reshape_8 = None + x_123 = x_122 + input_4 + x_122 = input_4 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_125 = x_124.permute(0, 2, 3, 1) + x_124 = None + v_13 = torch.var(x_125, dim=(-1,), keepdim=True) + add_22 = v_13 + 1e-06 + v_13 = None + rsqrt_13 = torch.rsqrt(add_22) + add_22 = None + x_126 = x_125 * rsqrt_13 + x_125 = rsqrt_13 = None + x_127 = ( + x_126 + * l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_130 = torch._C._nn.gelu(x_129) + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + reshape_9 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_134 = x_133.mul(reshape_9) + x_133 = reshape_9 = None + x_135 = x_134 + x_123 + x_134 = x_123 = None + x_136 = torch.nn.functional.adaptive_avg_pool2d(x_135, 1) + x_135 = None + x_137 = x_136.permute(0, 2, 3, 1) + x_136 = None + v_14 = torch.var(x_137, dim=(-1,), keepdim=True) + add_24 = v_14 + 1e-06 + v_14 = None + rsqrt_14 = torch.rsqrt(add_24) + add_24 = None + x_138 = x_137 * rsqrt_14 + x_137 = rsqrt_14 = None + x_139 = x_138 * l_self_modules_head_modules_norm_parameters_weight_ + x_138 = l_self_modules_head_modules_norm_parameters_weight_ = None + x_140 = x_139.permute(0, 3, 1, 2) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py new file mode 100644 index 000000000..1cab743f1 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py @@ -0,0 +1,974 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [64, 32, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt new file mode 100644 index 000000000..b73323104 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt @@ -0,0 +1 @@ +8c985b6584dd5c5878afa74dc799f3092db39d0e36501155abab738dbc8bae2d \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_meta.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_tensor_constraints.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py new file mode 100644 index 000000000..4282f6058 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py @@ -0,0 +1,1005 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stem_modules_3_parameters_weight_ = ( + L_self_modules_stem_modules_3_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch._C._nn.gelu(input_1) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_2 = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x = input_3.permute(0, 2, 3, 1) + input_3 = None + v = torch.var(x, dim=(-1,), keepdim=True) + add = v + 1e-06 + v = None + rsqrt = torch.rsqrt(add) + add = None + x_1 = x * rsqrt + x = rsqrt = None + x_2 = x_1 * l_self_modules_stem_modules_3_parameters_weight_ + x_1 = l_self_modules_stem_modules_3_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + x_4 = torch.conv2d( + x_3, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_5 = x_4.permute(0, 2, 3, 1) + x_4 = None + v_1 = torch.var(x_5, dim=(-1,), keepdim=True) + add_1 = v_1 + 1e-06 + v_1 = None + rsqrt_1 = torch.rsqrt(add_1) + add_1 = None + x_6 = x_5 * rsqrt_1 + x_5 = rsqrt_1 = None + x_7 = ( + x_6 + * l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_8 = x_7.permute(0, 3, 1, 2) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9) + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_14 = x_13.mul(reshape) + x_13 = reshape = None + x_15 = x_14 + x_3 + x_14 = x_3 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_17 = x_16.permute(0, 2, 3, 1) + x_16 = None + v_2 = torch.var(x_17, dim=(-1,), keepdim=True) + add_3 = v_2 + 1e-06 + v_2 = None + rsqrt_2 = torch.rsqrt(add_3) + add_3 = None + x_18 = x_17 * rsqrt_2 + x_17 = rsqrt_2 = None + x_19 = ( + x_18 + * l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_20 = x_19.permute(0, 3, 1, 2) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_22 = torch._C._nn.gelu(x_21) + x_21 = None + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_26 = x_25.mul(reshape_1) + x_25 = reshape_1 = None + x_27 = x_26 + x_15 + x_26 = x_15 = None + x_28 = x_27.permute(0, 2, 3, 1) + x_27 = None + v_3 = torch.var(x_28, dim=(-1,), keepdim=True) + add_5 = v_3 + 1e-06 + v_3 = None + rsqrt_3 = torch.rsqrt(add_5) + add_5 = None + x_29 = x_28 * rsqrt_3 + x_28 = rsqrt_3 = None + x_30 = ( + x_29 + * l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + ) + x_29 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + input_4 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_32 = torch.conv2d( + input_4, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_33 = x_32.permute(0, 2, 3, 1) + x_32 = None + v_4 = torch.var(x_33, dim=(-1,), keepdim=True) + add_6 = v_4 + 1e-06 + v_4 = None + rsqrt_4 = torch.rsqrt(add_6) + add_6 = None + x_34 = x_33 * rsqrt_4 + x_33 = rsqrt_4 = None + x_35 = ( + x_34 + * l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_36 = x_35.permute(0, 3, 1, 2) + x_35 = None + x_37 = torch.conv2d( + x_36, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_36 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_38 = torch._C._nn.gelu(x_37) + x_37 = None + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_41 = torch.nn.functional.dropout(x_40, 0.0, False, False) + x_40 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_42 = x_41.mul(reshape_2) + x_41 = reshape_2 = None + x_43 = x_42 + input_4 + x_42 = input_4 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_45 = x_44.permute(0, 2, 3, 1) + x_44 = None + v_5 = torch.var(x_45, dim=(-1,), keepdim=True) + add_8 = v_5 + 1e-06 + v_5 = None + rsqrt_5 = torch.rsqrt(add_8) + add_8 = None + x_46 = x_45 * rsqrt_5 + x_45 = rsqrt_5 = None + x_47 = ( + x_46 + * l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_50 = torch._C._nn.gelu(x_49) + x_49 = None + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_54 = x_53.mul(reshape_3) + x_53 = reshape_3 = None + x_55 = x_54 + x_43 + x_54 = x_43 = None + x_56 = x_55.permute(0, 2, 3, 1) + x_55 = None + v_6 = torch.var(x_56, dim=(-1,), keepdim=True) + add_10 = v_6 + 1e-06 + v_6 = None + rsqrt_6 = torch.rsqrt(add_10) + add_10 = None + x_57 = x_56 * rsqrt_6 + x_56 = rsqrt_6 = None + x_58 = ( + x_57 + * l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + ) + x_57 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = (None) + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + input_5 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_60 = torch.conv2d( + input_5, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_61 = x_60.permute(0, 2, 3, 1) + x_60 = None + v_7 = torch.var(x_61, dim=(-1,), keepdim=True) + add_11 = v_7 + 1e-06 + v_7 = None + rsqrt_7 = torch.rsqrt(add_11) + add_11 = None + x_62 = x_61 * rsqrt_7 + x_61 = rsqrt_7 = None + x_63 = ( + x_62 + * l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_4) + x_69 = reshape_4 = None + x_71 = x_70 + input_5 + x_70 = input_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + v_8 = torch.var(x_73, dim=(-1,), keepdim=True) + add_13 = v_8 + 1e-06 + v_8 = None + rsqrt_8 = torch.rsqrt(add_13) + add_13 = None + x_74 = x_73 * rsqrt_8 + x_73 = rsqrt_8 = None + x_75 = ( + x_74 + * l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_76 = x_75.permute(0, 3, 1, 2) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_78 = torch._C._nn.gelu(x_77) + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_82 = x_81.mul(reshape_5) + x_81 = reshape_5 = None + x_83 = x_82 + x_71 + x_82 = x_71 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_85 = x_84.permute(0, 2, 3, 1) + x_84 = None + v_9 = torch.var(x_85, dim=(-1,), keepdim=True) + add_15 = v_9 + 1e-06 + v_9 = None + rsqrt_9 = torch.rsqrt(add_15) + add_15 = None + x_86 = x_85 * rsqrt_9 + x_85 = rsqrt_9 = None + x_87 = ( + x_86 + * l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = (None) + x_88 = x_87.permute(0, 3, 1, 2) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_94 = x_93.mul(reshape_6) + x_93 = reshape_6 = None + x_95 = x_94 + x_83 + x_94 = x_83 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_97 = x_96.permute(0, 2, 3, 1) + x_96 = None + v_10 = torch.var(x_97, dim=(-1,), keepdim=True) + add_17 = v_10 + 1e-06 + v_10 = None + rsqrt_10 = torch.rsqrt(add_17) + add_17 = None + x_98 = x_97 * rsqrt_10 + x_97 = rsqrt_10 = None + x_99 = ( + x_98 + * l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_7) + x_105 = reshape_7 = None + x_107 = x_106 + x_95 + x_106 = x_95 = None + x_108 = x_107.permute(0, 2, 3, 1) + x_107 = None + v_11 = torch.var(x_108, dim=(-1,), keepdim=True) + add_19 = v_11 + 1e-06 + v_11 = None + rsqrt_11 = torch.rsqrt(add_19) + add_19 = None + x_109 = x_108 * rsqrt_11 + x_108 = rsqrt_11 = None + x_110 = ( + x_109 + * l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + ) + x_109 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + input_6 = torch.conv2d( + x_111, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_112 = torch.conv2d( + input_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_113 = x_112.permute(0, 2, 3, 1) + x_112 = None + v_12 = torch.var(x_113, dim=(-1,), keepdim=True) + add_20 = v_12 + 1e-06 + v_12 = None + rsqrt_12 = torch.rsqrt(add_20) + add_20 = None + x_114 = x_113 * rsqrt_12 + x_113 = rsqrt_12 = None + x_115 = ( + x_114 + * l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_114 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_118 = torch._C._nn.gelu(x_117) + x_117 = None + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + reshape_8 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_122 = x_121.mul(reshape_8) + x_121 = reshape_8 = None + x_123 = x_122 + input_6 + x_122 = input_6 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_125 = x_124.permute(0, 2, 3, 1) + x_124 = None + v_13 = torch.var(x_125, dim=(-1,), keepdim=True) + add_22 = v_13 + 1e-06 + v_13 = None + rsqrt_13 = torch.rsqrt(add_22) + add_22 = None + x_126 = x_125 * rsqrt_13 + x_125 = rsqrt_13 = None + x_127 = ( + x_126 + * l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_130 = torch._C._nn.gelu(x_129) + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + reshape_9 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_134 = x_133.mul(reshape_9) + x_133 = reshape_9 = None + x_135 = x_134 + x_123 + x_134 = x_123 = None + x_136 = torch.nn.functional.adaptive_avg_pool2d(x_135, 1) + x_135 = None + x_137 = x_136.permute(0, 2, 3, 1) + x_136 = None + v_14 = torch.var(x_137, dim=(-1,), keepdim=True) + add_24 = v_14 + 1e-06 + v_14 = None + rsqrt_14 = torch.rsqrt(add_24) + add_24 = None + x_138 = x_137 * rsqrt_14 + x_137 = rsqrt_14 = None + x_139 = x_138 * l_self_modules_head_modules_norm_parameters_weight_ + x_138 = l_self_modules_head_modules_norm_parameters_weight_ = None + x_140 = x_139.permute(0, 3, 1, 2) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py new file mode 100644 index 000000000..ee89ada53 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py @@ -0,0 +1,994 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_3_parameters_weight_: + name = "L_self_modules_stem_modules_3_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [64, 32, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_atto.fcmae/graph_hash.txt b/samples/timm/convnextv2_atto.fcmae/graph_hash.txt new file mode 100644 index 000000000..22926c008 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/graph_hash.txt @@ -0,0 +1 @@ +3cf0175b6d1f5c006dfd17d29027c4b6912846eec375b1a975876378f099af57 \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae/graph_net.json b/samples/timm/convnextv2_atto.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae/input_meta.py b/samples/timm/convnextv2_atto.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_atto.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_atto.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_atto.fcmae/model.py b/samples/timm/convnextv2_atto.fcmae/model.py new file mode 100644 index 000000000..b6f5fd7a9 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_atto.fcmae/weight_meta.py b/samples/timm/convnextv2_atto.fcmae/weight_meta.py new file mode 100644 index 000000000..68ee77ec0 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [40, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..24c226b77 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +c66466ed362e91b7ab9dec5c38bc4e70d2d78e515479441349935b2bc807f770 \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..26d24b3d1 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..f11ac5c7b --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [40, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_base.fcmae/graph_hash.txt b/samples/timm/convnextv2_base.fcmae/graph_hash.txt new file mode 100644 index 000000000..108b252cd --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/graph_hash.txt @@ -0,0 +1 @@ +8d3afeab77d8b825e8fe1abe687ffb77e6f066e13c4b04211471aad7ecf62c65 \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae/graph_net.json b/samples/timm/convnextv2_base.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae/input_meta.py b/samples/timm/convnextv2_base.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_base.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_base.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_base.fcmae/model.py b/samples/timm/convnextv2_base.fcmae/model.py new file mode 100644 index 000000000..c9f2785f8 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (128,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1024,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_base.fcmae/weight_meta.py b/samples/timm/convnextv2_base.fcmae/weight_meta.py new file mode 100644 index 000000000..c4cd9b38d --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..743b4867d --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +8d87ff876e26add5d5c71dcfb9bc70367e838d0b8e8ac6cb2562fd7661d035a5 \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..09c969a05 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (128,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1024,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..d29845ffb --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_femto.fcmae/graph_hash.txt b/samples/timm/convnextv2_femto.fcmae/graph_hash.txt new file mode 100644 index 000000000..e454cfc2b --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/graph_hash.txt @@ -0,0 +1 @@ +e5108a1f4678e444be04e203895b2668c3c7af06626cfc8faa2497b31f31b758 \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae/graph_net.json b/samples/timm/convnextv2_femto.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae/input_meta.py b/samples/timm/convnextv2_femto.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_femto.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_femto.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_femto.fcmae/model.py b/samples/timm/convnextv2_femto.fcmae/model.py new file mode 100644 index 000000000..fdb6c1927 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_femto.fcmae/weight_meta.py b/samples/timm/convnextv2_femto.fcmae/weight_meta.py new file mode 100644 index 000000000..fe65b3560 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [48, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..e7bfa26e2 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +4a937f17830161ff371fbc0820658beeedd7e940c7d349c7a86885d72baab3a6 \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..8cff47d96 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..634c1166e --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [48, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_huge.fcmae/graph_hash.txt b/samples/timm/convnextv2_huge.fcmae/graph_hash.txt new file mode 100644 index 000000000..23cdc31b0 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/graph_hash.txt @@ -0,0 +1 @@ +8fcb4693ac7ae2f805b5c736406cfc3dfd8b7c5eeffaea4e083ab3f65af787d5 \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae/graph_net.json b/samples/timm/convnextv2_huge.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae/input_meta.py b/samples/timm/convnextv2_huge.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_huge.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_huge.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_huge.fcmae/model.py b/samples/timm/convnextv2_huge.fcmae/model.py new file mode 100644 index 000000000..b98186968 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (352,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (352,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (704,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1408,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2816,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_huge.fcmae/weight_meta.py b/samples/timm/convnextv2_huge.fcmae/weight_meta.py new file mode 100644 index 000000000..ae4f9db92 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [352, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [704, 352, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1408, 704, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2816, 1408, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..f32efd31e --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +f89d8e3e398c7a58eb6fce1d1432881adf546a3af3016fdb63062b361da41a1e \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..f79548035 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (352,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (352,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (704,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1408,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2816,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..809be42e6 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [352, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [704, 352, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1408, 704, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2816, 1408, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_large.fcmae/graph_hash.txt b/samples/timm/convnextv2_large.fcmae/graph_hash.txt new file mode 100644 index 000000000..e31486c4f --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/graph_hash.txt @@ -0,0 +1 @@ +5edc6a6e18554e542bf2d9176814b8693f8f0f277af6cc827ca95f027f217354 \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae/graph_net.json b/samples/timm/convnextv2_large.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae/input_meta.py b/samples/timm/convnextv2_large.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_large.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_large.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_large.fcmae/model.py b/samples/timm/convnextv2_large.fcmae/model.py new file mode 100644 index 000000000..dae73df9e --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_large.fcmae/weight_meta.py b/samples/timm/convnextv2_large.fcmae/weight_meta.py new file mode 100644 index 000000000..c743c91ea --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..ca248166d --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +a959225c800ea702edb90763031e8d806ff70463f142397e37c3ad05ece4f96d \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..d78965baf --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..cedfe3724 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.232 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_nano.fcmae/graph_hash.txt b/samples/timm/convnextv2_nano.fcmae/graph_hash.txt new file mode 100644 index 000000000..acc1eacb1 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/graph_hash.txt @@ -0,0 +1 @@ +914c586df7bf0728ff0403e9e64ba9ace425ac173539baa29c2289106bd509d9 \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae/graph_net.json b/samples/timm/convnextv2_nano.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae/input_meta.py b/samples/timm/convnextv2_nano.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_nano.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_nano.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_nano.fcmae/model.py b/samples/timm/convnextv2_nano.fcmae/model.py new file mode 100644 index 000000000..2f84a1f38 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/model.py @@ -0,0 +1,1487 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_g_10 = x_125.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_125 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_126 = x_125 + addcmul_10 + x_125 = addcmul_10 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_g_11 = x_136.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_136 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_137 = x_136 + addcmul_11 + x_136 = addcmul_11 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_g_12 = x_150.norm(p=2, dim=(2, 3), keepdim=True) + mean_12 = x_g_12.mean(dim=1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_150 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_151 = x_150 + addcmul_12 + x_150 = addcmul_12 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_g_13 = x_161.norm(p=2, dim=(2, 3), keepdim=True) + mean_13 = x_g_13.mean(dim=1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_161 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_162 = x_161 + addcmul_13 + x_161 = addcmul_13 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + return (x_171,) diff --git a/samples/timm/convnextv2_nano.fcmae/weight_meta.py b/samples/timm/convnextv2_nano.fcmae/weight_meta.py new file mode 100644 index 000000000..db1b85b2b --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/weight_meta.py @@ -0,0 +1,1610 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..da2fc1fbf --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +b8181c9f9ff27ad7cb072ba4161f271d2e379ae487acc6a9fac1a9cc78ee1b70 \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..81a65077f --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py @@ -0,0 +1,1503 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_g_10 = x_125.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_125 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_126 = x_125 + addcmul_10 + x_125 = addcmul_10 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_g_11 = x_136.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_136 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_137 = x_136 + addcmul_11 + x_136 = addcmul_11 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_g_12 = x_150.norm(p=2, dim=(2, 3), keepdim=True) + mean_12 = x_g_12.mean(dim=1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_150 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_151 = x_150 + addcmul_12 + x_150 = addcmul_12 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_g_13 = x_161.norm(p=2, dim=(2, 3), keepdim=True) + mean_13 = x_g_13.mean(dim=1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_161 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_162 = x_161 + addcmul_13 + x_161 = addcmul_13 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..13baa05d4 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1630 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_pico.fcmae/graph_hash.txt b/samples/timm/convnextv2_pico.fcmae/graph_hash.txt new file mode 100644 index 000000000..b62217ee3 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/graph_hash.txt @@ -0,0 +1 @@ +69e21bb20c51b1d85583aa5f82c6408fd0bd4b948662d296dd39624afcc86010 \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae/graph_net.json b/samples/timm/convnextv2_pico.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae/input_meta.py b/samples/timm/convnextv2_pico.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_pico.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_pico.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_pico.fcmae/model.py b/samples/timm/convnextv2_pico.fcmae/model.py new file mode 100644 index 000000000..24ec0ea94 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_pico.fcmae/weight_meta.py b/samples/timm/convnextv2_pico.fcmae/weight_meta.py new file mode 100644 index 000000000..2a3d47306 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..3f8200a58 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +069e5e4f030b7e114c60364f33ab05ca64cb2970bb19b122e93fc7528c1c2c86 \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..67c81bec9 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..ee956f7d1 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt b/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt new file mode 100644 index 000000000..8b14ecd9e --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt @@ -0,0 +1 @@ +768dce85976b59285690f156b20646da6537dbcca7eb13625f756f057c303bea \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae/graph_net.json b/samples/timm/convnextv2_tiny.fcmae/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae/input_meta.py b/samples/timm/convnextv2_tiny.fcmae/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_tiny.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_tiny.fcmae/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_tiny.fcmae/model.py b/samples/timm/convnextv2_tiny.fcmae/model.py new file mode 100644 index 000000000..9df891fad --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/model.py @@ -0,0 +1,1719 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_g_15 = x_182.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_182 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_183 = x_182 + addcmul_15 + x_182 = addcmul_15 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + x_186 = x_185.permute(0, 3, 1, 2) + x_185 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_g_16 = x_193.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_193 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_194 = x_193 + addcmul_16 + x_193 = addcmul_16 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_g_17 = x_204.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_204 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_205 = x_204 + addcmul_17 + x_204 = addcmul_17 = None + x_206 = torch._C._nn.linear( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + x_208 = x_207.permute(0, 3, 1, 2) + x_207 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + return (x_215,) diff --git a/samples/timm/convnextv2_tiny.fcmae/weight_meta.py b/samples/timm/convnextv2_tiny.fcmae/weight_meta.py new file mode 100644 index 000000000..e9058b644 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/weight_meta.py @@ -0,0 +1,2010 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 000000000..4a54d453e --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +51358064ff58e02a77b4420c059759367a1fd15f6dac8988e275cc60c5a314b4 \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py new file mode 100644 index 000000000..0f3568244 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py @@ -0,0 +1,1735 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_g_15 = x_182.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_182 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_183 = x_182 + addcmul_15 + x_182 = addcmul_15 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + x_186 = x_185.permute(0, 3, 1, 2) + x_185 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_g_16 = x_193.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_193 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_194 = x_193 + addcmul_16 + x_193 = addcmul_16 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_g_17 = x_204.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_204 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_205 = x_204 + addcmul_17 + x_204 = addcmul_17 = None + x_206 = torch._C._nn.linear( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + x_208 = x_207.permute(0, 3, 1, 2) + x_207 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 000000000..b7de4f9be --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,2030 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_240.in1k/graph_hash.txt b/samples/timm/crossvit_15_240.in1k/graph_hash.txt new file mode 100644 index 000000000..e8b59ad2a --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +3a12e5809ccc0e1fdd5f7c10b900aadef0cd78c86eb87f88a87ef9b393136a85 \ No newline at end of file diff --git a/samples/timm/crossvit_15_240.in1k/graph_net.json b/samples/timm/crossvit_15_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_240.in1k/input_meta.py b/samples/timm/crossvit_15_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_240.in1k/model.py b/samples/timm/crossvit_15_240.in1k/model.py new file mode 100644 index 000000000..b48e5952e --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/model.py @@ -0,0 +1,2733 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + getitem_34 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_68[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_36), dim=1) + input_3 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 197, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 197, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_69 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_70 = torch._C._nn.linear( + x_69, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_69 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_71 = torch.nn.functional.dropout(x_70, 0.0, False, False) + x_70 = None + x_72 = getitem_37 + x_71 + getitem_37 = x_71 = None + getitem_39 = x_72[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_72 = None + input_7 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_40), dim=1) + input_9 = getitem_40 = None + getitem_41 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_41), dim=1) + input_6 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 401, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 401, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_73 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = getitem_42 + x_75 + getitem_42 = x_75 = None + getitem_44 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_76 = None + input_10 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_68 = None + tmp_3 = torch.cat((input_12, getitem_45), dim=1) + input_12 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 401, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 401, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_1 + x_80 + tmp_1 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_3 + x_91 + tmp_3 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 384) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 384) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + getitem_64 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_142[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_66), dim=1) + input_15 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 197, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_143 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = getitem_67 + x_145 + getitem_67 = x_145 = None + getitem_69 = x_146[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_146 = None + input_19 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_70), dim=1) + input_21 = getitem_70 = None + getitem_71 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_87 = None + tmp_6 = torch.cat((input_18, getitem_71), dim=1) + input_18 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 401, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 401, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_147 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = getitem_72 + x_149 + getitem_72 = x_149 = None + getitem_74 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_150 = None + input_22 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_142 = None + tmp_7 = torch.cat((input_24, getitem_75), dim=1) + input_24 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 401, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 401, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = tmp_5 + x_154 + tmp_5 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = tmp_7 + x_165 + tmp_7 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_183, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 384) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = x_183 + x_187 + x_183 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 384) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 384) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + getitem_94 = x_161[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_216[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_96), dim=1) + input_27 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 197, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_217 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_218 = torch._C._nn.linear( + x_217, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_217 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_219 = torch.nn.functional.dropout(x_218, 0.0, False, False) + x_218 = None + x_220 = getitem_97 + x_219 + getitem_97 = x_219 = None + getitem_99 = x_220[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_220 = None + input_31 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_100), dim=1) + input_33 = getitem_100 = None + getitem_101 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_161 = None + tmp_10 = torch.cat((input_30, getitem_101), dim=1) + input_30 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 401, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 401, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_221 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_221 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = getitem_102 + x_223 + getitem_102 = x_223 = None + getitem_104 = x_224[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_224 = None + input_34 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_216 = None + tmp_11 = torch.cat((input_36, getitem_105), dim=1) + input_36 = getitem_105 = None + x_225 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_226 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_227 = x_225[(slice(None, None, None), 0)] + x_225 = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + dropout_68 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + dropout_69 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_229 = torch.mean(stack, dim=0) + stack = None + return (x_229,) diff --git a/samples/timm/crossvit_15_240.in1k/weight_meta.py b/samples/timm/crossvit_15_240.in1k/weight_meta.py new file mode 100644 index 000000000..ac649f426 --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/weight_meta.py @@ -0,0 +1,3408 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [192, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt new file mode 100644 index 000000000..82cac6e4e --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +12e4a17846003b8fccb9f45487096cd775aab14f58f5b1129f1bf44433e25f1e \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_15_dagger_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_dagger_240.in1k/model.py b/samples/timm/crossvit_15_dagger_240.in1k/model.py new file mode 100644 index 000000000..de3b64c5c --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/model.py @@ -0,0 +1,2809 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + getitem_34 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_68[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_36), dim=1) + input_13 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 197, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 197, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_69 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_70 = torch._C._nn.linear( + x_69, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_69 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_71 = torch.nn.functional.dropout(x_70, 0.0, False, False) + x_70 = None + x_72 = getitem_37 + x_71 + getitem_37 = x_71 = None + getitem_39 = x_72[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_72 = None + input_17 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_40), dim=1) + input_19 = getitem_40 = None + getitem_41 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_41), dim=1) + input_16 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 401, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 401, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_73 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = getitem_42 + x_75 + getitem_42 = x_75 = None + getitem_44 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_76 = None + input_20 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_68 = None + tmp_3 = torch.cat((input_22, getitem_45), dim=1) + input_22 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 401, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 401, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_1 + x_80 + tmp_1 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_3 + x_91 + tmp_3 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 384) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 384) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + getitem_64 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_142[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_66), dim=1) + input_25 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 197, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_143 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = getitem_67 + x_145 + getitem_67 = x_145 = None + getitem_69 = x_146[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_146 = None + input_29 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_70), dim=1) + input_31 = getitem_70 = None + getitem_71 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_87 = None + tmp_6 = torch.cat((input_28, getitem_71), dim=1) + input_28 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 401, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 401, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_147 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = getitem_72 + x_149 + getitem_72 = x_149 = None + getitem_74 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_150 = None + input_32 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_142 = None + tmp_7 = torch.cat((input_34, getitem_75), dim=1) + input_34 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 401, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 401, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = tmp_5 + x_154 + tmp_5 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = tmp_7 + x_165 + tmp_7 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_183, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 384) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = x_183 + x_187 + x_183 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 384) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 384) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + getitem_94 = x_161[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_216[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_96), dim=1) + input_37 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 197, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_217 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_218 = torch._C._nn.linear( + x_217, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_217 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_219 = torch.nn.functional.dropout(x_218, 0.0, False, False) + x_218 = None + x_220 = getitem_97 + x_219 + getitem_97 = x_219 = None + getitem_99 = x_220[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_220 = None + input_41 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_100), dim=1) + input_43 = getitem_100 = None + getitem_101 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_161 = None + tmp_10 = torch.cat((input_40, getitem_101), dim=1) + input_40 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 401, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 401, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_221 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_221 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = getitem_102 + x_223 + getitem_102 = x_223 = None + getitem_104 = x_224[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_224 = None + input_44 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_216 = None + tmp_11 = torch.cat((input_46, getitem_105), dim=1) + input_46 = getitem_105 = None + x_225 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_226 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_227 = x_225[(slice(None, None, None), 0)] + x_225 = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + dropout_68 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + dropout_69 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_229 = torch.mean(stack, dim=0) + stack = None + return (x_229,) diff --git a/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py new file mode 100644 index 000000000..175c745a8 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py @@ -0,0 +1,3512 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [48, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [96, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt b/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt new file mode 100644 index 000000000..49fae451f --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt @@ -0,0 +1 @@ +66d5c1cd52f9c8e484bc9cce56fd0cfdb5cd0a275d6b9efd65469dda69dd7143 \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json b/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_408.in1k/input_meta.py b/samples/timm/crossvit_15_dagger_408.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_dagger_408.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_dagger_408.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_15_dagger_408.in1k/model.py b/samples/timm/crossvit_15_dagger_408.in1k/model.py new file mode 100644 index 000000000..b648a7ea0 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/model.py @@ -0,0 +1,2814 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(408, 408), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + x_2 = torch.nn.functional.interpolate( + l_x_, size=(384, 384), mode="bicubic", align_corners=False + ) + l_x_ = None + input_6 = torch.conv2d( + x_2, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x_2 = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_3 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_3), dim=1) + cls_tokens_1 = x_3 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 1157, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_4 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_4.transpose(1, 2) + x_4 = None + x_5 = transpose_2.reshape(1, 1157, 192) + transpose_2 = None + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_7 = torch.nn.functional.dropout(x_6, 0.0, False, False) + x_6 = None + x_8 = x__2 + x_7 + x__2 = x_7 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_8, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_9 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9, approximate="none") + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + x_14 = x_8 + x_13 + x_8 = x_13 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 577, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_15 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_15.transpose(1, 2) + x_15 = None + x_16 = transpose_3.reshape(1, 577, 384) + transpose_3 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_16 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x__5 + x_18 + x__5 = x_18 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_19, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_20 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_21 = torch._C._nn.gelu(x_20, approximate="none") + x_20 = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = torch._C._nn.linear( + x_22, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_22 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_24 = torch.nn.functional.dropout(x_23, 0.0, False, False) + x_23 = None + x_25 = x_19 + x_24 + x_19 = x_24 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_25, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 577, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_26 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_26.transpose(1, 2) + x_26 = None + x_27 = transpose_4.reshape(1, 577, 384) + transpose_4 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_27 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_25 + x_29 + x_25 = x_29 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_31 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_32 = torch._C._nn.gelu(x_31, approximate="none") + x_31 = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_33 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 577, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_37 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_37.transpose(1, 2) + x_37 = None + x_38 = transpose_5.reshape(1, 577, 384) + transpose_5 = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_36 + x_40 + x_36 = x_40 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_41, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42, approximate="none") + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_41 + x_46 + x_41 = x_46 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_47, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 577, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_48 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_48.transpose(1, 2) + x_48 = None + x_49 = transpose_6.reshape(1, 577, 384) + transpose_6 = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_49 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = x_47 + x_51 + x_47 = x_51 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_52, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53, approximate="none") + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_52 + x_57 + x_52 = x_57 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_58, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 577, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_59 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_59.transpose(1, 2) + x_59 = None + x_60 = transpose_7.reshape(1, 577, 384) + transpose_7 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_60 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_58 + x_62 + x_58 = x_62 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_63, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_63 + x_68 + x_63 = x_68 = None + getitem_34 = x_14[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_69[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_69[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_36), dim=1) + input_13 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 577, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 577, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_70 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = getitem_37 + x_72 + getitem_37 = x_72 = None + getitem_39 = x_73[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_73 = None + input_17 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_40), dim=1) + input_19 = getitem_40 = None + getitem_41 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_14 = None + tmp_2 = torch.cat((input_16, getitem_41), dim=1) + input_16 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 1157, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 1157, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_74 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_74 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = getitem_42 + x_76 + getitem_42 = x_76 = None + getitem_44 = x_77[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_77 = None + input_20 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_69[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_69 = None + tmp_3 = torch.cat((input_22, getitem_45), dim=1) + input_22 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 1157, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_78 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_78.transpose(1, 2) + x_78 = None + x_79 = transpose_12.reshape(1, 1157, 192) + transpose_12 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_79 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = tmp_1 + x_81 + tmp_1 = x_81 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_83 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_84 = torch._C._nn.gelu(x_83, approximate="none") + x_83 = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_85 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 577, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_89 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_89.transpose(1, 2) + x_89 = None + x_90 = transpose_13.reshape(1, 577, 384) + transpose_13 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_90 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = tmp_3 + x_92 + tmp_3 = x_92 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_93, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_94 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_95 = torch._C._nn.gelu(x_94, approximate="none") + x_94 = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + x_97 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_96 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_98 = torch.nn.functional.dropout(x_97, 0.0, False, False) + x_97 = None + x_99 = x_93 + x_98 + x_93 = x_98 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_99, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 577, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_100 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_100.transpose(1, 2) + x_100 = None + x_101 = transpose_14.reshape(1, 577, 384) + transpose_14 = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_101 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = x_99 + x_103 + x_99 = x_103 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_104, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_105 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_106 = torch._C._nn.gelu(x_105, approximate="none") + x_105 = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = torch._C._nn.linear( + x_107, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_107 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_109 = torch.nn.functional.dropout(x_108, 0.0, False, False) + x_108 = None + x_110 = x_104 + x_109 + x_104 = x_109 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_110, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 577, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_15.reshape(1, 577, 384) + transpose_15 = None + x_113 = torch._C._nn.linear( + x_112, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_112 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = x_110 + x_114 + x_110 = x_114 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_115, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_116 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_115 + x_120 + x_115 = x_120 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 577, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_122 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_122.transpose(1, 2) + x_122 = None + x_123 = transpose_16.reshape(1, 577, 384) + transpose_16 = None + x_124 = torch._C._nn.linear( + x_123, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_123 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_121 + x_125 + x_121 = x_125 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_126, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_127 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_128 = torch._C._nn.gelu(x_127, approximate="none") + x_127 = None + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_129 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_126 + x_131 + x_126 = x_131 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_132, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 577, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_133 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_133.transpose(1, 2) + x_133 = None + x_134 = transpose_17.reshape(1, 577, 384) + transpose_17 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = x_132 + x_136 + x_132 = x_136 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_137, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_138 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_137 + x_142 + x_137 = x_142 = None + getitem_64 = x_88[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_143[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_143[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_66), dim=1) + input_25 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 577, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 577, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_144 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = getitem_67 + x_146 + getitem_67 = x_146 = None + getitem_69 = x_147[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_147 = None + input_29 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_88[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_70), dim=1) + input_31 = getitem_70 = None + getitem_71 = x_88[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_88 = None + tmp_6 = torch.cat((input_28, getitem_71), dim=1) + input_28 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 1157, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 1157, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_148 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_149 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_148 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = getitem_72 + x_150 + getitem_72 = x_150 = None + getitem_74 = x_151[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_151 = None + input_32 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_143[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_143 = None + tmp_7 = torch.cat((input_34, getitem_75), dim=1) + input_34 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 1157, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_152 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_152.transpose(1, 2) + x_152 = None + x_153 = transpose_22.reshape(1, 1157, 192) + transpose_22 = None + x_154 = torch._C._nn.linear( + x_153, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_153 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) + x_154 = None + x_156 = tmp_5 + x_155 + tmp_5 = x_155 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_156, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_157 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_158 = torch._C._nn.gelu(x_157, approximate="none") + x_157 = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = torch._C._nn.linear( + x_159, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_159 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = x_156 + x_161 + x_156 = x_161 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 577, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_163 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_163.transpose(1, 2) + x_163 = None + x_164 = transpose_23.reshape(1, 577, 384) + transpose_23 = None + x_165 = torch._C._nn.linear( + x_164, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_164 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + x_167 = tmp_7 + x_166 + tmp_7 = x_166 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_167, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_168 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + x_173 = x_167 + x_172 + x_167 = x_172 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_173, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 577, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_174 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_174.transpose(1, 2) + x_174 = None + x_175 = transpose_24.reshape(1, 577, 384) + transpose_24 = None + x_176 = torch._C._nn.linear( + x_175, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_175 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = x_173 + x_177 + x_173 = x_177 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_178, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_179 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_180 = torch._C._nn.gelu(x_179, approximate="none") + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_178 + x_183 + x_178 = x_183 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_184, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 577, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_185 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_185.transpose(1, 2) + x_185 = None + x_186 = transpose_25.reshape(1, 577, 384) + transpose_25 = None + x_187 = torch._C._nn.linear( + x_186, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_186 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = x_184 + x_188 + x_184 = x_188 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_189, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_190 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_191 = torch._C._nn.gelu(x_190, approximate="none") + x_190 = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_192 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = x_189 + x_194 + x_189 = x_194 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_195, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 577, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_196 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_196.transpose(1, 2) + x_196 = None + x_197 = transpose_26.reshape(1, 577, 384) + transpose_26 = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_197 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = x_195 + x_199 + x_195 = x_199 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_200, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_201 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_202 = torch._C._nn.gelu(x_201, approximate="none") + x_201 = None + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_200 + x_205 + x_200 = x_205 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_206, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 577, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_207 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_207.transpose(1, 2) + x_207 = None + x_208 = transpose_27.reshape(1, 577, 384) + transpose_27 = None + x_209 = torch._C._nn.linear( + x_208, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_208 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_206 + x_210 + x_206 = x_210 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_211, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_212 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_213 = torch._C._nn.gelu(x_212, approximate="none") + x_212 = None + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = x_211 + x_216 + x_211 = x_216 = None + getitem_94 = x_162[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_217[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_217[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_96), dim=1) + input_37 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 577, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 577, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_218 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = getitem_97 + x_220 + getitem_97 = x_220 = None + getitem_99 = x_221[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_221 = None + input_41 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_162[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_100), dim=1) + input_43 = getitem_100 = None + getitem_101 = x_162[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_162 = None + tmp_10 = torch.cat((input_40, getitem_101), dim=1) + input_40 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 1157, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 1157, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_222 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_222 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = getitem_102 + x_224 + getitem_102 = x_224 = None + getitem_104 = x_225[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_225 = None + input_44 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_217[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_217 = None + tmp_11 = torch.cat((input_46, getitem_105), dim=1) + input_46 = getitem_105 = None + x_226 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_227 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + x_229 = x_227[(slice(None, None, None), 0)] + x_227 = None + dropout_68 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + dropout_69 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_230 = torch.mean(stack, dim=0) + stack = None + return (x_230,) diff --git a/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py b/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py new file mode 100644 index 000000000..980a0d7bc --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py @@ -0,0 +1,3522 @@ +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [48, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [96, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 1157, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 577, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_240.in1k/graph_hash.txt b/samples/timm/crossvit_18_240.in1k/graph_hash.txt new file mode 100644 index 000000000..b98987842 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +191d9cd47d0904f532b3160647b8b1caaa2bf2c6c8c73b9a6b64dbe3c5d6e6c2 \ No newline at end of file diff --git a/samples/timm/crossvit_18_240.in1k/graph_net.json b/samples/timm/crossvit_18_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_240.in1k/input_meta.py b/samples/timm/crossvit_18_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_240.in1k/model.py b/samples/timm/crossvit_18_240.in1k/model.py new file mode 100644 index 000000000..b5b7817d6 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/model.py @@ -0,0 +1,3015 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 224) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 448) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 448) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 448) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 448) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 448) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_68, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_69 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_69.transpose(1, 2) + x_69 = None + x_70 = transpose_8.reshape(1, 197, 448) + transpose_8 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = x_68 + x_72 + x_68 = x_72 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_73, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_74 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_75 = torch._C._nn.gelu(x_74, approximate="none") + x_74 = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_73 + x_78 + x_73 = x_78 = None + getitem_37 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_79[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_39), dim=1) + input_3 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 197, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 197, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_80 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_80 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = getitem_40 + x_82 + getitem_40 = x_82 = None + getitem_42 = x_83[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_83 = None + input_7 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_43), dim=1) + input_9 = getitem_43 = None + getitem_44 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_44), dim=1) + input_6 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 401, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 401, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_84 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = getitem_45 + x_86 + getitem_45 = x_86 = None + getitem_47 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_87 = None + input_10 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_79 = None + tmp_3 = torch.cat((input_12, getitem_48), dim=1) + input_12 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 401, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 401, 224) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_1 + x_91 + tmp_1 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 448) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = tmp_3 + x_102 + tmp_3 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 448) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 448) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 448) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_142, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 197, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_143 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_143.transpose(1, 2) + x_143 = None + x_144 = transpose_18.reshape(1, 197, 448) + transpose_18 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_142 + x_146 + x_142 = x_146 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_147, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_148 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_153, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_154 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_154.transpose(1, 2) + x_154 = None + x_155 = transpose_19.reshape(1, 197, 448) + transpose_19 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = x_153 + x_157 + x_153 = x_157 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_158, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_159 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159, approximate="none") + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_161 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + getitem_70 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_164[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_72), dim=1) + input_15 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 197, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 197, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_165 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_165 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_167 = torch.nn.functional.dropout(x_166, 0.0, False, False) + x_166 = None + x_168 = getitem_73 + x_167 + getitem_73 = x_167 = None + getitem_75 = x_168[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_168 = None + input_19 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_76), dim=1) + input_21 = getitem_76 = None + getitem_77 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_6 = torch.cat((input_18, getitem_77), dim=1) + input_18 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 401, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 401, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_169 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = getitem_78 + x_171 + getitem_78 = x_171 = None + getitem_80 = x_172[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_172 = None + input_22 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_164 = None + tmp_7 = torch.cat((input_24, getitem_81), dim=1) + input_24 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 401, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 401, 224) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = tmp_5 + x_176 + tmp_5 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 448) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = tmp_7 + x_187 + tmp_7 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 448) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 448) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_216, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 197, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_217 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_217.transpose(1, 2) + x_217 = None + x_218 = transpose_28.reshape(1, 197, 448) + transpose_28 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_216 + x_220 + x_216 = x_220 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_221, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_222 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_223 = torch._C._nn.gelu(x_222, approximate="none") + x_222 = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_221 + x_226 + x_221 = x_226 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_227, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_228 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_228.transpose(1, 2) + x_228 = None + x_229 = transpose_29.reshape(1, 197, 448) + transpose_29 = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_229 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_231 = torch.nn.functional.dropout(x_230, 0.0, False, False) + x_230 = None + x_232 = x_227 + x_231 + x_227 = x_231 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_232, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_233 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_232 + x_237 + x_232 = x_237 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_238, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 197, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_239 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_239.transpose(1, 2) + x_239 = None + x_240 = transpose_30.reshape(1, 197, 448) + transpose_30 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_240 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = x_238 + x_242 + x_238 = x_242 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_243, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_244 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_243 + x_248 + x_243 = x_248 = None + getitem_103 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_249[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_105), dim=1) + input_27 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 197, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 197, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_250 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_251 = torch._C._nn.linear( + x_250, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_250 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + x_253 = getitem_106 + x_252 + getitem_106 = x_252 = None + getitem_108 = x_253[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_253 = None + input_31 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_109), dim=1) + input_33 = getitem_109 = None + getitem_110 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_10 = torch.cat((input_30, getitem_110), dim=1) + input_30 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 401, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 401, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_254 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = getitem_111 + x_256 + getitem_111 = x_256 = None + getitem_113 = x_257[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_257 = None + input_34 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_249 = None + tmp_11 = torch.cat((input_36, getitem_114), dim=1) + input_36 = getitem_114 = None + x_258 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_259 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_260 = x_258[(slice(None, None, None), 0)] + x_258 = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + dropout_77 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + dropout_78 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_262 = torch.mean(stack, dim=0) + stack = None + return (x_262,) diff --git a/samples/timm/crossvit_18_240.in1k/weight_meta.py b/samples/timm/crossvit_18_240.in1k/weight_meta.py new file mode 100644 index 000000000..f8c3efb49 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/weight_meta.py @@ -0,0 +1,3768 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [224, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [448, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt new file mode 100644 index 000000000..e1d20be9a --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +38cd8450bce14ad3d7db0b79ba42ef2cc211152c220e9615c9ecadec74e45cff \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_18_dagger_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_dagger_240.in1k/model.py b/samples/timm/crossvit_18_dagger_240.in1k/model.py new file mode 100644 index 000000000..98891b26b --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 224) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 448) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 448) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 448) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 448) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 448) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_68, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_69 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_69.transpose(1, 2) + x_69 = None + x_70 = transpose_8.reshape(1, 197, 448) + transpose_8 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = x_68 + x_72 + x_68 = x_72 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_73, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_74 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_75 = torch._C._nn.gelu(x_74, approximate="none") + x_74 = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_73 + x_78 + x_73 = x_78 = None + getitem_37 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_79[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_39), dim=1) + input_13 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 197, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 197, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_80 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_80 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = getitem_40 + x_82 + getitem_40 = x_82 = None + getitem_42 = x_83[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_83 = None + input_17 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_43), dim=1) + input_19 = getitem_43 = None + getitem_44 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_44), dim=1) + input_16 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 401, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 401, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_84 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = getitem_45 + x_86 + getitem_45 = x_86 = None + getitem_47 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_87 = None + input_20 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_79 = None + tmp_3 = torch.cat((input_22, getitem_48), dim=1) + input_22 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 401, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 401, 224) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_1 + x_91 + tmp_1 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 448) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = tmp_3 + x_102 + tmp_3 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 448) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 448) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 448) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_142, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 197, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_143 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_143.transpose(1, 2) + x_143 = None + x_144 = transpose_18.reshape(1, 197, 448) + transpose_18 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_142 + x_146 + x_142 = x_146 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_147, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_148 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_153, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_154 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_154.transpose(1, 2) + x_154 = None + x_155 = transpose_19.reshape(1, 197, 448) + transpose_19 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = x_153 + x_157 + x_153 = x_157 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_158, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_159 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159, approximate="none") + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_161 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + getitem_70 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_164[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_72), dim=1) + input_25 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 197, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 197, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_165 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_165 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_167 = torch.nn.functional.dropout(x_166, 0.0, False, False) + x_166 = None + x_168 = getitem_73 + x_167 + getitem_73 = x_167 = None + getitem_75 = x_168[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_168 = None + input_29 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_76), dim=1) + input_31 = getitem_76 = None + getitem_77 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_6 = torch.cat((input_28, getitem_77), dim=1) + input_28 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 401, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 401, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_169 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = getitem_78 + x_171 + getitem_78 = x_171 = None + getitem_80 = x_172[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_172 = None + input_32 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_164 = None + tmp_7 = torch.cat((input_34, getitem_81), dim=1) + input_34 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 401, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 401, 224) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = tmp_5 + x_176 + tmp_5 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 448) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = tmp_7 + x_187 + tmp_7 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 448) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 448) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_216, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 197, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_217 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_217.transpose(1, 2) + x_217 = None + x_218 = transpose_28.reshape(1, 197, 448) + transpose_28 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_216 + x_220 + x_216 = x_220 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_221, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_222 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_223 = torch._C._nn.gelu(x_222, approximate="none") + x_222 = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_221 + x_226 + x_221 = x_226 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_227, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_228 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_228.transpose(1, 2) + x_228 = None + x_229 = transpose_29.reshape(1, 197, 448) + transpose_29 = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_229 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_231 = torch.nn.functional.dropout(x_230, 0.0, False, False) + x_230 = None + x_232 = x_227 + x_231 + x_227 = x_231 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_232, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_233 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_232 + x_237 + x_232 = x_237 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_238, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 197, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_239 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_239.transpose(1, 2) + x_239 = None + x_240 = transpose_30.reshape(1, 197, 448) + transpose_30 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_240 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = x_238 + x_242 + x_238 = x_242 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_243, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_244 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_243 + x_248 + x_243 = x_248 = None + getitem_103 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_249[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_105), dim=1) + input_37 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 197, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 197, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_250 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_251 = torch._C._nn.linear( + x_250, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_250 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + x_253 = getitem_106 + x_252 + getitem_106 = x_252 = None + getitem_108 = x_253[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_253 = None + input_41 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_109), dim=1) + input_43 = getitem_109 = None + getitem_110 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_10 = torch.cat((input_40, getitem_110), dim=1) + input_40 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 401, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 401, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_254 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = getitem_111 + x_256 + getitem_111 = x_256 = None + getitem_113 = x_257[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_257 = None + input_44 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_249 = None + tmp_11 = torch.cat((input_46, getitem_114), dim=1) + input_46 = getitem_114 = None + x_258 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_259 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_260 = x_258[(slice(None, None, None), 0)] + x_258 = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + dropout_77 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + dropout_78 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_262 = torch.mean(stack, dim=0) + stack = None + return (x_262,) diff --git a/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py new file mode 100644 index 000000000..564b4082d --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py @@ -0,0 +1,3872 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [56, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [56] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [112, 56, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [112, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.011 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [448, 224, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.012 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt b/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt new file mode 100644 index 000000000..98580fb36 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt @@ -0,0 +1 @@ +f22e958a02b95aa6cfc33f02b15734dc4dbdacecfe968467dedfb62f0b896cde \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json b/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_408.in1k/input_meta.py b/samples/timm/crossvit_18_dagger_408.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_dagger_408.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_dagger_408.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_18_dagger_408.in1k/model.py b/samples/timm/crossvit_18_dagger_408.in1k/model.py new file mode 100644 index 000000000..ea63f99b2 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/model.py @@ -0,0 +1,3096 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(408, 408), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + x_2 = torch.nn.functional.interpolate( + l_x_, size=(384, 384), mode="bicubic", align_corners=False + ) + l_x_ = None + input_6 = torch.conv2d( + x_2, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x_2 = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_3 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_3), dim=1) + cls_tokens_1 = x_3 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 1157, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_4 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_4.transpose(1, 2) + x_4 = None + x_5 = transpose_2.reshape(1, 1157, 224) + transpose_2 = None + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_7 = torch.nn.functional.dropout(x_6, 0.0, False, False) + x_6 = None + x_8 = x__2 + x_7 + x__2 = x_7 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_8, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_9 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9, approximate="none") + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + x_14 = x_8 + x_13 + x_8 = x_13 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 577, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_15 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_15.transpose(1, 2) + x_15 = None + x_16 = transpose_3.reshape(1, 577, 448) + transpose_3 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_16 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x__5 + x_18 + x__5 = x_18 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_19, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_20 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_21 = torch._C._nn.gelu(x_20, approximate="none") + x_20 = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = torch._C._nn.linear( + x_22, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_22 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_24 = torch.nn.functional.dropout(x_23, 0.0, False, False) + x_23 = None + x_25 = x_19 + x_24 + x_19 = x_24 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_25, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 577, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_26 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_26.transpose(1, 2) + x_26 = None + x_27 = transpose_4.reshape(1, 577, 448) + transpose_4 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_27 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_25 + x_29 + x_25 = x_29 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_30, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_31 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_32 = torch._C._nn.gelu(x_31, approximate="none") + x_31 = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_33 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_36, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 577, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_37 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_37.transpose(1, 2) + x_37 = None + x_38 = transpose_5.reshape(1, 577, 448) + transpose_5 = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_36 + x_40 + x_36 = x_40 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_41, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42, approximate="none") + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_41 + x_46 + x_41 = x_46 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_47, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 577, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_48 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_48.transpose(1, 2) + x_48 = None + x_49 = transpose_6.reshape(1, 577, 448) + transpose_6 = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_49 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = x_47 + x_51 + x_47 = x_51 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_52, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53, approximate="none") + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_52 + x_57 + x_52 = x_57 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_58, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 577, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_59 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_59.transpose(1, 2) + x_59 = None + x_60 = transpose_7.reshape(1, 577, 448) + transpose_7 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_60 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_58 + x_62 + x_58 = x_62 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_63, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_63 + x_68 + x_63 = x_68 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_69, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 577, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_70 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_70.transpose(1, 2) + x_70 = None + x_71 = transpose_8.reshape(1, 577, 448) + transpose_8 = None + x_72 = torch._C._nn.linear( + x_71, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_71 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_69 + x_73 + x_69 = x_73 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_74, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_75 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_76 = torch._C._nn.gelu(x_75, approximate="none") + x_75 = None + x_77 = torch.nn.functional.dropout(x_76, 0.0, False, False) + x_76 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_77 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_74 + x_79 + x_74 = x_79 = None + getitem_37 = x_14[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_80[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_80[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_39), dim=1) + input_13 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 577, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 577, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_81 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_81 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = getitem_40 + x_83 + getitem_40 = x_83 = None + getitem_42 = x_84[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_84 = None + input_17 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_43), dim=1) + input_19 = getitem_43 = None + getitem_44 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_14 = None + tmp_2 = torch.cat((input_16, getitem_44), dim=1) + input_16 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 1157, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 1157, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_85 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = getitem_45 + x_87 + getitem_45 = x_87 = None + getitem_47 = x_88[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_88 = None + input_20 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_80[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_80 = None + tmp_3 = torch.cat((input_22, getitem_48), dim=1) + input_22 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 1157, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_89 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_89.transpose(1, 2) + x_89 = None + x_90 = transpose_13.reshape(1, 1157, 224) + transpose_13 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_90 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = tmp_1 + x_92 + tmp_1 = x_92 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_93, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_94 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_95 = torch._C._nn.gelu(x_94, approximate="none") + x_94 = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + x_97 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_96 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_98 = torch.nn.functional.dropout(x_97, 0.0, False, False) + x_97 = None + x_99 = x_93 + x_98 + x_93 = x_98 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 577, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_100 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_100.transpose(1, 2) + x_100 = None + x_101 = transpose_14.reshape(1, 577, 448) + transpose_14 = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_101 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = tmp_3 + x_103 + tmp_3 = x_103 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_104, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_105 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_106 = torch._C._nn.gelu(x_105, approximate="none") + x_105 = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = torch._C._nn.linear( + x_107, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_107 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_109 = torch.nn.functional.dropout(x_108, 0.0, False, False) + x_108 = None + x_110 = x_104 + x_109 + x_104 = x_109 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_110, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 577, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_15.reshape(1, 577, 448) + transpose_15 = None + x_113 = torch._C._nn.linear( + x_112, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_112 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = x_110 + x_114 + x_110 = x_114 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_115, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_116 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_115 + x_120 + x_115 = x_120 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_121, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 577, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_122 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_122.transpose(1, 2) + x_122 = None + x_123 = transpose_16.reshape(1, 577, 448) + transpose_16 = None + x_124 = torch._C._nn.linear( + x_123, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_123 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_121 + x_125 + x_121 = x_125 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_126, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_127 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_128 = torch._C._nn.gelu(x_127, approximate="none") + x_127 = None + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_129 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_126 + x_131 + x_126 = x_131 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_132, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 577, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_133 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_133.transpose(1, 2) + x_133 = None + x_134 = transpose_17.reshape(1, 577, 448) + transpose_17 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = x_132 + x_136 + x_132 = x_136 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_137, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_138 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_137 + x_142 + x_137 = x_142 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_143, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 577, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_144 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_144.transpose(1, 2) + x_144 = None + x_145 = transpose_18.reshape(1, 577, 448) + transpose_18 = None + x_146 = torch._C._nn.linear( + x_145, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_145 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = x_143 + x_147 + x_143 = x_147 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_148, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_149 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_150 = torch._C._nn.gelu(x_149, approximate="none") + x_149 = None + x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) + x_150 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_148 + x_153 + x_148 = x_153 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_154, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 577, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_155 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_155.transpose(1, 2) + x_155 = None + x_156 = transpose_19.reshape(1, 577, 448) + transpose_19 = None + x_157 = torch._C._nn.linear( + x_156, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_156 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = x_154 + x_158 + x_154 = x_158 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_159, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_160 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_161 = torch._C._nn.gelu(x_160, approximate="none") + x_160 = None + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_162 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_159 + x_164 + x_159 = x_164 = None + getitem_70 = x_99[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_165[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_165[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_72), dim=1) + input_25 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 577, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 577, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_166 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_166 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = getitem_73 + x_168 + getitem_73 = x_168 = None + getitem_75 = x_169[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_169 = None + input_29 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_99[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_76), dim=1) + input_31 = getitem_76 = None + getitem_77 = x_99[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_99 = None + tmp_6 = torch.cat((input_28, getitem_77), dim=1) + input_28 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 1157, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 1157, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_170 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_170 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + x_173 = getitem_78 + x_172 + getitem_78 = x_172 = None + getitem_80 = x_173[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_173 = None + input_32 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_165[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_165 = None + tmp_7 = torch.cat((input_34, getitem_81), dim=1) + input_34 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 1157, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_174 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_174.transpose(1, 2) + x_174 = None + x_175 = transpose_24.reshape(1, 1157, 224) + transpose_24 = None + x_176 = torch._C._nn.linear( + x_175, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_175 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = tmp_5 + x_177 + tmp_5 = x_177 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_178, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_179 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_180 = torch._C._nn.gelu(x_179, approximate="none") + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_178 + x_183 + x_178 = x_183 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 577, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_185 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_185.transpose(1, 2) + x_185 = None + x_186 = transpose_25.reshape(1, 577, 448) + transpose_25 = None + x_187 = torch._C._nn.linear( + x_186, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_186 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = tmp_7 + x_188 + tmp_7 = x_188 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_189, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_190 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_191 = torch._C._nn.gelu(x_190, approximate="none") + x_190 = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_192 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = x_189 + x_194 + x_189 = x_194 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_195, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 577, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_196 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_196.transpose(1, 2) + x_196 = None + x_197 = transpose_26.reshape(1, 577, 448) + transpose_26 = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_197 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = x_195 + x_199 + x_195 = x_199 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_200, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_201 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_202 = torch._C._nn.gelu(x_201, approximate="none") + x_201 = None + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_200 + x_205 + x_200 = x_205 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_206, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 577, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_207 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_207.transpose(1, 2) + x_207 = None + x_208 = transpose_27.reshape(1, 577, 448) + transpose_27 = None + x_209 = torch._C._nn.linear( + x_208, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_208 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_206 + x_210 + x_206 = x_210 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_211, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_212 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_213 = torch._C._nn.gelu(x_212, approximate="none") + x_212 = None + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = x_211 + x_216 + x_211 = x_216 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_217, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 577, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_218 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_218.transpose(1, 2) + x_218 = None + x_219 = transpose_28.reshape(1, 577, 448) + transpose_28 = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_219 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_221 = torch.nn.functional.dropout(x_220, 0.0, False, False) + x_220 = None + x_222 = x_217 + x_221 + x_217 = x_221 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_222, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_223 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch._C._nn.linear( + x_225, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_225 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_222 + x_227 + x_222 = x_227 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_228, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 577, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_229 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_229.transpose(1, 2) + x_229 = None + x_230 = transpose_29.reshape(1, 577, 448) + transpose_29 = None + x_231 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_230 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + x_233 = x_228 + x_232 + x_228 = x_232 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_233, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_234 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_235 = torch._C._nn.gelu(x_234, approximate="none") + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_236 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = x_233 + x_238 + x_233 = x_238 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_239, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 577, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_30.reshape(1, 577, 448) + transpose_30 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_239 + x_243 + x_239 = x_243 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_244, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_245 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_246 = torch._C._nn.gelu(x_245, approximate="none") + x_245 = None + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = torch._C._nn.linear( + x_247, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_247 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = x_244 + x_249 + x_244 = x_249 = None + getitem_103 = x_184[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_250[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_250[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_105), dim=1) + input_37 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 577, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 577, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_251 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_252 = torch._C._nn.linear( + x_251, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_251 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_253 = torch.nn.functional.dropout(x_252, 0.0, False, False) + x_252 = None + x_254 = getitem_106 + x_253 + getitem_106 = x_253 = None + getitem_108 = x_254[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_254 = None + input_41 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_184[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_109), dim=1) + input_43 = getitem_109 = None + getitem_110 = x_184[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_184 = None + tmp_10 = torch.cat((input_40, getitem_110), dim=1) + input_40 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 1157, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 1157, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_255 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_256 = torch._C._nn.linear( + x_255, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_255 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_257 = torch.nn.functional.dropout(x_256, 0.0, False, False) + x_256 = None + x_258 = getitem_111 + x_257 + getitem_111 = x_257 = None + getitem_113 = x_258[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_258 = None + input_44 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_250[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_250 = None + tmp_11 = torch.cat((input_46, getitem_114), dim=1) + input_46 = getitem_114 = None + x_259 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_260 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + x_262 = x_260[(slice(None, None, None), 0)] + x_260 = None + dropout_77 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + dropout_78 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_263 = torch.mean(stack, dim=0) + stack = None + return (x_263,) diff --git a/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py b/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py new file mode 100644 index 000000000..1551a691e --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py @@ -0,0 +1,3882 @@ +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [56, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [56] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [112, 56, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 1157, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [112, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [448, 224, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 577, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_9_240.in1k/graph_hash.txt b/samples/timm/crossvit_9_240.in1k/graph_hash.txt new file mode 100644 index 000000000..132be6b21 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +7f4aa4bf0e6727cd9b1cbdaecbb0c382d588738468f7dd687ec41bcc79f65ffa \ No newline at end of file diff --git a/samples/timm/crossvit_9_240.in1k/graph_net.json b/samples/timm/crossvit_9_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_9_240.in1k/input_meta.py b/samples/timm/crossvit_9_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_9_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_9_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_9_240.in1k/model.py b/samples/timm/crossvit_9_240.in1k/model.py new file mode 100644 index 000000000..f84c380c7 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/model.py @@ -0,0 +1,2169 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 4, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 128) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 4, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 256) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 4, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 256) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 4, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 256) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + getitem_28 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_28, + (128,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_28 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_29 = x_46[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_29, + (256,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_29 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_30 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_30), dim=1) + input_3 = getitem_30 = None + getitem_31 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_10 = torch.nn.functional.layer_norm( + tmp, + (256,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_32 = layer_norm_10[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_18 = torch._C._nn.linear( + getitem_32, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_8 = linear_18.reshape(1, 1, 4, 64) + linear_18 = None + q_4 = reshape_8.permute(0, 2, 1, 3) + reshape_8 = None + linear_19 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_9 = linear_19.reshape(1, 197, 4, 64) + linear_19 = None + k_4 = reshape_9.permute(0, 2, 1, 3) + reshape_9 = None + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 4, 64) + linear_20 = None + v_4 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + transpose_6 = k_4.transpose(-2, -1) + k_4 = None + matmul = q_4 @ transpose_6 + q_4 = transpose_6 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_4 + attn_2 = v_4 = None + transpose_7 = matmul_1.transpose(1, 2) + matmul_1 = None + x_47 = transpose_7.reshape(1, 1, 256) + transpose_7 = None + x_48 = torch._C._nn.linear( + x_47, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_47 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_49 = torch.nn.functional.dropout(x_48, 0.0, False, False) + x_48 = None + x_50 = getitem_31 + x_49 + getitem_31 = x_49 = None + getitem_33 = x_50[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_50 = None + input_7 = torch.nn.functional.layer_norm( + getitem_33, + (256,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_33 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_34 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_34), dim=1) + input_9 = getitem_34 = None + getitem_35 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_35), dim=1) + input_6 = getitem_35 = None + getitem_36 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp_2, + (128,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_37 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_23 = torch._C._nn.linear( + getitem_37, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_23.reshape(1, 1, 4, 32) + linear_23 = None + q_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_24.reshape(1, 401, 4, 32) + linear_24 = None + k_5 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_25 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_25.reshape(1, 401, 4, 32) + linear_25 = None + v_5 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_5.transpose(-2, -1) + k_5 = None + matmul_2 = q_5 @ transpose_8 + q_5 = transpose_8 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_5 + attn_5 = v_5 = None + transpose_9 = matmul_3.transpose(1, 2) + matmul_3 = None + x_51 = transpose_9.reshape(1, 1, 128) + transpose_9 = None + x_52 = torch._C._nn.linear( + x_51, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_51 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = getitem_36 + x_53 + getitem_36 = x_53 = None + getitem_38 = x_54[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_54 = None + input_10 = torch.nn.functional.layer_norm( + getitem_38, + (128,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_46 = None + tmp_3 = torch.cat((input_12, getitem_39), dim=1) + input_12 = getitem_39 = None + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_1, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_16 = linear_28.reshape(1, 401, 3, 4, 32) + linear_28 = None + qkv_4 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_6 = unbind_4[0] + k_6 = unbind_4[1] + v_6 = unbind_4[2] + unbind_4 = None + x_55 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_10 = x_55.transpose(1, 2) + x_55 = None + x_56 = transpose_10.reshape(1, 401, 128) + transpose_10 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_56 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = tmp_1 + x_58 + tmp_1 = x_58 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_60 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_61 = torch._C._nn.gelu(x_60, approximate="none") + x_60 = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = x_59 + x_64 + x_59 = x_64 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_3, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 197, 3, 4, 64) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 197, 256) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_3 + x_69 + tmp_3 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 4, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 256) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = x_76 + x_80 + x_76 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 4, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 256) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + getitem_52 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_52, + (128,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_52 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_53 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_53, + (256,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_53 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_54 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_54), dim=1) + input_15 = getitem_54 = None + getitem_55 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_24 = torch.nn.functional.layer_norm( + tmp_4, + (256,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_56 = layer_norm_24[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_46 = torch._C._nn.linear( + getitem_56, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_56 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_24 = linear_46.reshape(1, 1, 4, 64) + linear_46 = None + q_10 = reshape_24.permute(0, 2, 1, 3) + reshape_24 = None + linear_47 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_25 = linear_47.reshape(1, 197, 4, 64) + linear_47 = None + k_10 = reshape_25.permute(0, 2, 1, 3) + reshape_25 = None + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 4, 64) + linear_48 = None + v_10 = reshape_26.permute(0, 2, 1, 3) + reshape_26 = None + transpose_14 = k_10.transpose(-2, -1) + k_10 = None + matmul_4 = q_10 @ transpose_14 + q_10 = transpose_14 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_10 + attn_8 = v_10 = None + transpose_15 = matmul_5.transpose(1, 2) + matmul_5 = None + x_99 = transpose_15.reshape(1, 1, 256) + transpose_15 = None + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_99 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_101 = torch.nn.functional.dropout(x_100, 0.0, False, False) + x_100 = None + x_102 = getitem_55 + x_101 + getitem_55 = x_101 = None + getitem_57 = x_102[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_102 = None + input_19 = torch.nn.functional.layer_norm( + getitem_57, + (256,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_57 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_58 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_58), dim=1) + input_21 = getitem_58 = None + getitem_59 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_65 = None + tmp_6 = torch.cat((input_18, getitem_59), dim=1) + input_18 = getitem_59 = None + getitem_60 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_26 = torch.nn.functional.layer_norm( + tmp_6, + (128,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_61 = layer_norm_26[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_51 = torch._C._nn.linear( + getitem_61, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_61 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_51.reshape(1, 1, 4, 32) + linear_51 = None + q_11 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_52.reshape(1, 401, 4, 32) + linear_52 = None + k_11 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_53 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_53.reshape(1, 401, 4, 32) + linear_53 = None + v_11 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_11.transpose(-2, -1) + k_11 = None + matmul_6 = q_11 @ transpose_16 + q_11 = transpose_16 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_11 + attn_11 = v_11 = None + transpose_17 = matmul_7.transpose(1, 2) + matmul_7 = None + x_103 = transpose_17.reshape(1, 1, 128) + transpose_17 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_103 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = getitem_60 + x_105 + getitem_60 = x_105 = None + getitem_62 = x_106[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_106 = None + input_22 = torch.nn.functional.layer_norm( + getitem_62, + (128,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_63 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_7 = torch.cat((input_24, getitem_63), dim=1) + input_24 = getitem_63 = None + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_5, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_56.reshape(1, 401, 3, 4, 32) + linear_56 = None + qkv_8 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_12 = unbind_8[0] + k_12 = unbind_8[1] + v_12 = unbind_8[2] + unbind_8 = None + x_107 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_18 = x_107.transpose(1, 2) + x_107 = None + x_108 = transpose_18.reshape(1, 401, 128) + transpose_18 = None + x_109 = torch._C._nn.linear( + x_108, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_108 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = tmp_5 + x_110 + tmp_5 = x_110 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_111, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_112 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112, approximate="none") + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_111 + x_116 + x_111 = x_116 = None + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_7, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_60.reshape(1, 197, 3, 4, 64) + linear_60 = None + qkv_9 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_13 = unbind_9[0] + k_13 = unbind_9[1] + v_13 = unbind_9[2] + unbind_9 = None + x_118 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_19 = x_118.transpose(1, 2) + x_118 = None + x_119 = transpose_19.reshape(1, 197, 256) + transpose_19 = None + x_120 = torch._C._nn.linear( + x_119, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_119 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + x_122 = tmp_7 + x_121 + tmp_7 = x_121 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_122, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_123 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123, approximate="none") + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_122 + x_127 + x_122 = x_127 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_128, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 197, 3, 4, 64) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 197, 256) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = x_128 + x_132 + x_128 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_139, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 4, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 256) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = x_139 + x_143 + x_139 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + getitem_76 = x_117[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_76, + (128,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_76 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_77 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_77, + (256,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_77 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_78 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_78), dim=1) + input_27 = getitem_78 = None + getitem_79 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_8, + (256,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_80 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_74 = torch._C._nn.linear( + getitem_80, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_80 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_74.reshape(1, 1, 4, 64) + linear_74 = None + q_16 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_75 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_75.reshape(1, 197, 4, 64) + linear_75 = None + k_16 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 4, 64) + linear_76 = None + v_16 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_16.transpose(-2, -1) + k_16 = None + matmul_8 = q_16 @ transpose_22 + q_16 = transpose_22 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_16 + attn_14 = v_16 = None + transpose_23 = matmul_9.transpose(1, 2) + matmul_9 = None + x_151 = transpose_23.reshape(1, 1, 256) + transpose_23 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = getitem_79 + x_153 + getitem_79 = x_153 = None + getitem_81 = x_154[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_154 = None + input_31 = torch.nn.functional.layer_norm( + getitem_81, + (256,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_81 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_82 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_82), dim=1) + input_33 = getitem_82 = None + getitem_83 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_117 = None + tmp_10 = torch.cat((input_30, getitem_83), dim=1) + input_30 = getitem_83 = None + getitem_84 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_10, + (128,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_85 = layer_norm_40[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_79 = torch._C._nn.linear( + getitem_85, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_44 = linear_79.reshape(1, 1, 4, 32) + linear_79 = None + q_17 = reshape_44.permute(0, 2, 1, 3) + reshape_44 = None + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_45 = linear_80.reshape(1, 401, 4, 32) + linear_80 = None + k_17 = reshape_45.permute(0, 2, 1, 3) + reshape_45 = None + linear_81 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_46 = linear_81.reshape(1, 401, 4, 32) + linear_81 = None + v_17 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + transpose_24 = k_17.transpose(-2, -1) + k_17 = None + matmul_10 = q_17 @ transpose_24 + q_17 = transpose_24 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_17 + attn_17 = v_17 = None + transpose_25 = matmul_11.transpose(1, 2) + matmul_11 = None + x_155 = transpose_25.reshape(1, 1, 128) + transpose_25 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = getitem_84 + x_157 + getitem_84 = x_157 = None + getitem_86 = x_158[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_158 = None + input_34 = torch.nn.functional.layer_norm( + getitem_86, + (128,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_150 = None + tmp_11 = torch.cat((input_36, getitem_87), dim=1) + input_36 = getitem_87 = None + x_159 = torch.nn.functional.layer_norm( + tmp_9, + (128,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_160 = torch.nn.functional.layer_norm( + tmp_11, + (256,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_161 = x_159[(slice(None, None, None), 0)] + x_159 = None + x_162 = x_160[(slice(None, None, None), 0)] + x_160 = None + dropout_50 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + dropout_51 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + linear_84 = torch._C._nn.linear( + dropout_50, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_50 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_85 = torch._C._nn.linear( + dropout_51, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_51 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_84, linear_85], dim=0) + linear_84 = linear_85 = None + x_163 = torch.mean(stack, dim=0) + stack = None + return (x_163,) diff --git a/samples/timm/crossvit_9_240.in1k/weight_meta.py b/samples/timm/crossvit_9_240.in1k/weight_meta.py new file mode 100644 index 000000000..86db4db2e --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/weight_meta.py @@ -0,0 +1,2688 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [128, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [256, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt new file mode 100644 index 000000000..c6b98d5f1 --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +e01655a005966ed579a9eeb52b1521d441377c3157438eb7f4877ec9c13d775a \ No newline at end of file diff --git a/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_9_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_9_dagger_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_9_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_9_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_9_dagger_240.in1k/model.py b/samples/timm/crossvit_9_dagger_240.in1k/model.py new file mode 100644 index 000000000..33c73d0aa --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/model.py @@ -0,0 +1,2245 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 4, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 128) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 4, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 256) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 4, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 256) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 4, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 256) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + getitem_28 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_28, + (128,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_28 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_29 = x_46[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_29, + (256,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_29 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_30 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_30), dim=1) + input_13 = getitem_30 = None + getitem_31 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_10 = torch.nn.functional.layer_norm( + tmp, + (256,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_32 = layer_norm_10[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_18 = torch._C._nn.linear( + getitem_32, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_8 = linear_18.reshape(1, 1, 4, 64) + linear_18 = None + q_4 = reshape_8.permute(0, 2, 1, 3) + reshape_8 = None + linear_19 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_9 = linear_19.reshape(1, 197, 4, 64) + linear_19 = None + k_4 = reshape_9.permute(0, 2, 1, 3) + reshape_9 = None + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 4, 64) + linear_20 = None + v_4 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + transpose_6 = k_4.transpose(-2, -1) + k_4 = None + matmul = q_4 @ transpose_6 + q_4 = transpose_6 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_4 + attn_2 = v_4 = None + transpose_7 = matmul_1.transpose(1, 2) + matmul_1 = None + x_47 = transpose_7.reshape(1, 1, 256) + transpose_7 = None + x_48 = torch._C._nn.linear( + x_47, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_47 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_49 = torch.nn.functional.dropout(x_48, 0.0, False, False) + x_48 = None + x_50 = getitem_31 + x_49 + getitem_31 = x_49 = None + getitem_33 = x_50[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_50 = None + input_17 = torch.nn.functional.layer_norm( + getitem_33, + (256,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_33 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_34 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_34), dim=1) + input_19 = getitem_34 = None + getitem_35 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_35), dim=1) + input_16 = getitem_35 = None + getitem_36 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp_2, + (128,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_37 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_23 = torch._C._nn.linear( + getitem_37, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_23.reshape(1, 1, 4, 32) + linear_23 = None + q_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_24.reshape(1, 401, 4, 32) + linear_24 = None + k_5 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_25 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_25.reshape(1, 401, 4, 32) + linear_25 = None + v_5 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_5.transpose(-2, -1) + k_5 = None + matmul_2 = q_5 @ transpose_8 + q_5 = transpose_8 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_5 + attn_5 = v_5 = None + transpose_9 = matmul_3.transpose(1, 2) + matmul_3 = None + x_51 = transpose_9.reshape(1, 1, 128) + transpose_9 = None + x_52 = torch._C._nn.linear( + x_51, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_51 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = getitem_36 + x_53 + getitem_36 = x_53 = None + getitem_38 = x_54[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_54 = None + input_20 = torch.nn.functional.layer_norm( + getitem_38, + (128,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_46 = None + tmp_3 = torch.cat((input_22, getitem_39), dim=1) + input_22 = getitem_39 = None + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_1, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_16 = linear_28.reshape(1, 401, 3, 4, 32) + linear_28 = None + qkv_4 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_6 = unbind_4[0] + k_6 = unbind_4[1] + v_6 = unbind_4[2] + unbind_4 = None + x_55 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_10 = x_55.transpose(1, 2) + x_55 = None + x_56 = transpose_10.reshape(1, 401, 128) + transpose_10 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_56 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = tmp_1 + x_58 + tmp_1 = x_58 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_60 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_61 = torch._C._nn.gelu(x_60, approximate="none") + x_60 = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = x_59 + x_64 + x_59 = x_64 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_3, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 197, 3, 4, 64) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 197, 256) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_3 + x_69 + tmp_3 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 4, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 256) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = x_76 + x_80 + x_76 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 4, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 256) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + getitem_52 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_52, + (128,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_52 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_53 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_53, + (256,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_53 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_54 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_54), dim=1) + input_25 = getitem_54 = None + getitem_55 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_24 = torch.nn.functional.layer_norm( + tmp_4, + (256,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_56 = layer_norm_24[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_46 = torch._C._nn.linear( + getitem_56, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_56 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_24 = linear_46.reshape(1, 1, 4, 64) + linear_46 = None + q_10 = reshape_24.permute(0, 2, 1, 3) + reshape_24 = None + linear_47 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_25 = linear_47.reshape(1, 197, 4, 64) + linear_47 = None + k_10 = reshape_25.permute(0, 2, 1, 3) + reshape_25 = None + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 4, 64) + linear_48 = None + v_10 = reshape_26.permute(0, 2, 1, 3) + reshape_26 = None + transpose_14 = k_10.transpose(-2, -1) + k_10 = None + matmul_4 = q_10 @ transpose_14 + q_10 = transpose_14 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_10 + attn_8 = v_10 = None + transpose_15 = matmul_5.transpose(1, 2) + matmul_5 = None + x_99 = transpose_15.reshape(1, 1, 256) + transpose_15 = None + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_99 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_101 = torch.nn.functional.dropout(x_100, 0.0, False, False) + x_100 = None + x_102 = getitem_55 + x_101 + getitem_55 = x_101 = None + getitem_57 = x_102[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_102 = None + input_29 = torch.nn.functional.layer_norm( + getitem_57, + (256,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_57 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_58 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_58), dim=1) + input_31 = getitem_58 = None + getitem_59 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_65 = None + tmp_6 = torch.cat((input_28, getitem_59), dim=1) + input_28 = getitem_59 = None + getitem_60 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_26 = torch.nn.functional.layer_norm( + tmp_6, + (128,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_61 = layer_norm_26[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_51 = torch._C._nn.linear( + getitem_61, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_61 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_51.reshape(1, 1, 4, 32) + linear_51 = None + q_11 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_52.reshape(1, 401, 4, 32) + linear_52 = None + k_11 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_53 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_53.reshape(1, 401, 4, 32) + linear_53 = None + v_11 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_11.transpose(-2, -1) + k_11 = None + matmul_6 = q_11 @ transpose_16 + q_11 = transpose_16 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_11 + attn_11 = v_11 = None + transpose_17 = matmul_7.transpose(1, 2) + matmul_7 = None + x_103 = transpose_17.reshape(1, 1, 128) + transpose_17 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_103 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = getitem_60 + x_105 + getitem_60 = x_105 = None + getitem_62 = x_106[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_106 = None + input_32 = torch.nn.functional.layer_norm( + getitem_62, + (128,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_63 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_7 = torch.cat((input_34, getitem_63), dim=1) + input_34 = getitem_63 = None + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_5, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_56.reshape(1, 401, 3, 4, 32) + linear_56 = None + qkv_8 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_12 = unbind_8[0] + k_12 = unbind_8[1] + v_12 = unbind_8[2] + unbind_8 = None + x_107 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_18 = x_107.transpose(1, 2) + x_107 = None + x_108 = transpose_18.reshape(1, 401, 128) + transpose_18 = None + x_109 = torch._C._nn.linear( + x_108, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_108 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = tmp_5 + x_110 + tmp_5 = x_110 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_111, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_112 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112, approximate="none") + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_111 + x_116 + x_111 = x_116 = None + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_7, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_60.reshape(1, 197, 3, 4, 64) + linear_60 = None + qkv_9 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_13 = unbind_9[0] + k_13 = unbind_9[1] + v_13 = unbind_9[2] + unbind_9 = None + x_118 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_19 = x_118.transpose(1, 2) + x_118 = None + x_119 = transpose_19.reshape(1, 197, 256) + transpose_19 = None + x_120 = torch._C._nn.linear( + x_119, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_119 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + x_122 = tmp_7 + x_121 + tmp_7 = x_121 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_122, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_123 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123, approximate="none") + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_122 + x_127 + x_122 = x_127 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_128, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 197, 3, 4, 64) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 197, 256) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = x_128 + x_132 + x_128 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_139, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 4, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 256) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = x_139 + x_143 + x_139 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + getitem_76 = x_117[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_76, + (128,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_76 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_77 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_77, + (256,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_77 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_78 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_78), dim=1) + input_37 = getitem_78 = None + getitem_79 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_8, + (256,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_80 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_74 = torch._C._nn.linear( + getitem_80, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_80 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_74.reshape(1, 1, 4, 64) + linear_74 = None + q_16 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_75 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_75.reshape(1, 197, 4, 64) + linear_75 = None + k_16 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 4, 64) + linear_76 = None + v_16 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_16.transpose(-2, -1) + k_16 = None + matmul_8 = q_16 @ transpose_22 + q_16 = transpose_22 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_16 + attn_14 = v_16 = None + transpose_23 = matmul_9.transpose(1, 2) + matmul_9 = None + x_151 = transpose_23.reshape(1, 1, 256) + transpose_23 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = getitem_79 + x_153 + getitem_79 = x_153 = None + getitem_81 = x_154[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_154 = None + input_41 = torch.nn.functional.layer_norm( + getitem_81, + (256,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_81 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_82 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_82), dim=1) + input_43 = getitem_82 = None + getitem_83 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_117 = None + tmp_10 = torch.cat((input_40, getitem_83), dim=1) + input_40 = getitem_83 = None + getitem_84 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_10, + (128,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_85 = layer_norm_40[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_79 = torch._C._nn.linear( + getitem_85, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_44 = linear_79.reshape(1, 1, 4, 32) + linear_79 = None + q_17 = reshape_44.permute(0, 2, 1, 3) + reshape_44 = None + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_45 = linear_80.reshape(1, 401, 4, 32) + linear_80 = None + k_17 = reshape_45.permute(0, 2, 1, 3) + reshape_45 = None + linear_81 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_46 = linear_81.reshape(1, 401, 4, 32) + linear_81 = None + v_17 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + transpose_24 = k_17.transpose(-2, -1) + k_17 = None + matmul_10 = q_17 @ transpose_24 + q_17 = transpose_24 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_17 + attn_17 = v_17 = None + transpose_25 = matmul_11.transpose(1, 2) + matmul_11 = None + x_155 = transpose_25.reshape(1, 1, 128) + transpose_25 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = getitem_84 + x_157 + getitem_84 = x_157 = None + getitem_86 = x_158[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_158 = None + input_44 = torch.nn.functional.layer_norm( + getitem_86, + (128,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_150 = None + tmp_11 = torch.cat((input_46, getitem_87), dim=1) + input_46 = getitem_87 = None + x_159 = torch.nn.functional.layer_norm( + tmp_9, + (128,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_160 = torch.nn.functional.layer_norm( + tmp_11, + (256,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_161 = x_159[(slice(None, None, None), 0)] + x_159 = None + x_162 = x_160[(slice(None, None, None), 0)] + x_160 = None + dropout_50 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + dropout_51 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + linear_84 = torch._C._nn.linear( + dropout_50, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_50 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_85 = torch._C._nn.linear( + dropout_51, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_51 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_84, linear_85], dim=0) + linear_84 = linear_85 = None + x_163 = torch.mean(stack, dim=0) + stack = None + return (x_163,) diff --git a/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py new file mode 100644 index 000000000..2441199f5 --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py @@ -0,0 +1,2792 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [32, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.014 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_base_240.in1k/graph_hash.txt b/samples/timm/crossvit_base_240.in1k/graph_hash.txt new file mode 100644 index 000000000..66cadf2f2 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +e0627bcc7072a0e864ffc9d221b879ec6bdfca1250ff5d9551b6586a43c220dd \ No newline at end of file diff --git a/samples/timm/crossvit_base_240.in1k/graph_net.json b/samples/timm/crossvit_base_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_base_240.in1k/input_meta.py b/samples/timm/crossvit_base_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_base_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_base_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_base_240.in1k/model.py b/samples/timm/crossvit_base_240.in1k/model.py new file mode 100644 index 000000000..4ff3bd330 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 12, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 384) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (768,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (768,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 12, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 12, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 12, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 768) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (768,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 12, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 12, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 12, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 384) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 12, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 384) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 768) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 768) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 12, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 768) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (768,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (768,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 12, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 12, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 12, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 768) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (768,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 12, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 12, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 12, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 12, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 384) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 12, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 768) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 12, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 768) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 12, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 768) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 12, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 768) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (768,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (768,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 12, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 12, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 12, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 768) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (768,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 12, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 12, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 12, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 384) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (384,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (768,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_base_240.in1k/weight_meta.py b/samples/timm/crossvit_base_240.in1k/weight_meta.py new file mode 100644 index 000000000..7f6f7c456 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [384, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_small_240.in1k/graph_hash.txt b/samples/timm/crossvit_small_240.in1k/graph_hash.txt new file mode 100644 index 000000000..76341e912 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +4e02fa5c93460e927ac885c21327cd13c87d12faffd9a3573df684cde66ec654 \ No newline at end of file diff --git a/samples/timm/crossvit_small_240.in1k/graph_net.json b/samples/timm/crossvit_small_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_small_240.in1k/input_meta.py b/samples/timm/crossvit_small_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_small_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_small_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_small_240.in1k/model.py b/samples/timm/crossvit_small_240.in1k/model.py new file mode 100644 index 000000000..f8c01c9aa --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 6, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 6, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 6, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 384) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 6, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 6, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 6, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 192) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 6, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 192) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 6, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 6, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 6, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 384) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 6, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 6, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 6, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 192) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 6, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 192) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 6, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 384) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 6, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 384) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 6, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 6, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 6, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 384) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 6, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 6, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 6, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 192) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_small_240.in1k/weight_meta.py b/samples/timm/crossvit_small_240.in1k/weight_meta.py new file mode 100644 index 000000000..5fc073d13 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [192, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt b/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt new file mode 100644 index 000000000..b1530ba6b --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +63cd9da64f85a0be1d1bcf7b2ed6260cfecbf0e6393ac01b3b155379bce98355 \ No newline at end of file diff --git a/samples/timm/crossvit_tiny_240.in1k/graph_net.json b/samples/timm/crossvit_tiny_240.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_tiny_240.in1k/input_meta.py b/samples/timm/crossvit_tiny_240.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_tiny_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_tiny_240.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/crossvit_tiny_240.in1k/model.py b/samples/timm/crossvit_tiny_240.in1k/model.py new file mode 100644 index 000000000..131b83597 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (96,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 3, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 96) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (96,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 192) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 192) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 192) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 192) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (96,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 3, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 3, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 192) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (96,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 3, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 3, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 3, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 96) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (96,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (96,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 3, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 96) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (96,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 3, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 3, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 192) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 3, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 192) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 3, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 192) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (96,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 3, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 3, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 192) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (96,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 3, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 3, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 3, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 96) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (96,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (96,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 3, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 96) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (96,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 3, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 192) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 3, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 3, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 192) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 3, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 192) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (96,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 3, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 3, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 192) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (96,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 3, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 3, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 3, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 96) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (96,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (96,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (192,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_tiny_240.in1k/weight_meta.py b/samples/timm/crossvit_tiny_240.in1k/weight_meta.py new file mode 100644 index 000000000..1013e8ace --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [96, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..41a4d772c --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +b1b5d407a9ce4756c0f5f058742f0f1c7d9451a8b87c22a969d7db3f00854a06 \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py new file mode 100644 index 000000000..67fa432fe --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py @@ -0,0 +1,1992 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(64, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.silu(x_24, inplace=True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + cat = torch.cat([x_29, x2], dim=1) + x_29 = x2 = None + x_30 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.silu(x_31, inplace=True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_32 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + split_1 = x_38.split(128, dim=1) + x_38 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_39 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x1_1 + x_44 = x1_1 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x_45 + x_51 = x_45 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = x_58 + x_52 + x_58 = x_52 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_61 = torch.nn.functional.batch_norm( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_60 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_62 = torch.nn.functional.silu(x_61, inplace=True) + x_61 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = x_65 + x_59 + x_65 = x_59 = None + x_67 = torch.conv2d( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_68 = torch.nn.functional.batch_norm( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_69 = torch.nn.functional.silu(x_68, inplace=True) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = x_72 + x_66 + x_72 = x_66 = None + x_74 = torch.conv2d( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_75 = torch.nn.functional.batch_norm( + x_74, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_74 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_76 = torch.nn.functional.silu(x_75, inplace=True) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = x_79 + x_73 + x_79 = x_73 = None + cat_1 = torch.cat([x_80, x2_1], dim=1) + x_80 = x2_1 = None + x_81 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_83 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + split_2 = x_89.split(256, dim=1) + x_89 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_90 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = x_95 + x1_2 + x_95 = x1_2 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = x_102 + x_96 + x_102 = x_96 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = x_109 + x_103 + x_109 = x_103 = None + x_111 = torch.conv2d( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_112 = torch.nn.functional.batch_norm( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_113 = torch.nn.functional.silu(x_112, inplace=True) + x_112 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x_110 + x_116 = x_110 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = x_123 + x_117 + x_123 = x_117 = None + x_125 = torch.conv2d( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_126 = torch.nn.functional.batch_norm( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_127 = torch.nn.functional.silu(x_126, inplace=True) + x_126 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = x_130 + x_124 + x_130 = x_124 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_133 = torch.nn.functional.batch_norm( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_134 = torch.nn.functional.silu(x_133, inplace=True) + x_133 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = x_137 + x_131 + x_137 = x_131 = None + x_139 = torch.conv2d( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_140 = torch.nn.functional.batch_norm( + x_139, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_139 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_141 = torch.nn.functional.silu(x_140, inplace=True) + x_140 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = x_144 + x_138 + x_144 = x_138 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_148 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = x_151 + x_145 + x_151 = x_145 = None + cat_2 = torch.cat([x_152, x2_2], dim=1) + x_152 = x2_2 = None + x_153 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_154 = torch.nn.functional.batch_norm( + x_153, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_155 = torch.nn.functional.silu(x_154, inplace=True) + x_154 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_155 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + split_3 = x_161.split(512, dim=1) + x_161 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_162 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.silu(x_166, inplace=True) + x_166 = None + x_168 = x_167 + x1_3 + x_167 = x1_3 = None + x_169 = torch.conv2d( + x_168, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_170 = torch.nn.functional.batch_norm( + x_169, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_169 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_171 = torch.nn.functional.silu(x_170, inplace=True) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.silu(x_173, inplace=True) + x_173 = None + x_175 = x_174 + x_168 + x_174 = x_168 = None + x_176 = torch.conv2d( + x_175, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_177 = torch.nn.functional.batch_norm( + x_176, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_176 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_178 = torch.nn.functional.silu(x_177, inplace=True) + x_177 = None + x_179 = torch.conv2d( + x_178, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_181 = torch.nn.functional.silu(x_180, inplace=True) + x_180 = None + x_182 = x_181 + x_175 + x_181 = x_175 = None + cat_3 = torch.cat([x_182, x2_3], dim=1) + x_182 = x2_3 = None + x_183 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + x_186 = torch.nn.functional.adaptive_avg_pool2d(x_185, 1) + x_185 = None + x_187 = x_186.flatten(1, -1) + x_186 = None + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_188 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_189,) diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..f035c3b0c --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,2812 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.180 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..a339e5e87 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +f3fa74260ceba665ad095e7ef6c381c42711ac3f3ca201cd4825c0f522f7dbde \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py new file mode 100644 index 000000000..75b43c5d7 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py @@ -0,0 +1,1502 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(48, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + cat = torch.cat([x_22, x2], dim=1) + x_22 = x2 = None + x_23 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.silu(x_24, inplace=True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + split_1 = x_31.split(96, dim=1) + x_31 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_32 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = x_37 + x1_1 + x_37 = x1_1 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x_38 + x_44 = x_38 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x_45 + x_51 = x_45 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = x_58 + x_52 + x_58 = x_52 = None + cat_1 = torch.cat([x_59, x2_1], dim=1) + x_59 = x2_1 = None + x_60 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_61 = torch.nn.functional.batch_norm( + x_60, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_60 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_62 = torch.nn.functional.silu(x_61, inplace=True) + x_61 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_62 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + split_2 = x_68.split(192, dim=1) + x_68 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_69 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = x_74 + x1_2 + x_74 = x1_2 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.silu(x_77, inplace=True) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = x_81 + x_75 + x_81 = x_75 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_84 = torch.nn.functional.batch_norm( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_85 = torch.nn.functional.silu(x_84, inplace=True) + x_84 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = x_88 + x_82 + x_88 = x_82 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = x_95 + x_89 + x_95 = x_89 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = x_102 + x_96 + x_102 = x_96 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = x_109 + x_103 + x_109 = x_103 = None + cat_2 = torch.cat([x_110, x2_2], dim=1) + x_110 = x2_2 = None + x_111 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + x_111, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_111 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_113 = torch.nn.functional.silu(x_112, inplace=True) + x_112 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + split_3 = x_119.split(384, dim=1) + x_119 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_120 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.silu(x_121, inplace=True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.silu(x_124, inplace=True) + x_124 = None + x_126 = x_125 + x1_3 + x_125 = x1_3 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.silu(x_128, inplace=True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + cat_3 = torch.cat([x_133, x2_3], dim=1) + x_133 = x2_3 = None + x_134 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_137 = torch.nn.functional.adaptive_avg_pool2d(x_136, 1) + x_136 = None + x_138 = x_137.flatten(1, -1) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_139 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_140,) diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..3ce5e8ad9 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py @@ -0,0 +1,2112 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [96, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.207 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.204 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [768, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt new file mode 100644 index 000000000..7cbd7875c --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt @@ -0,0 +1 @@ +338be4333e6e92fd045ec42c48b512f53c2984df5171d281b13dfe16fe00d35a \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_meta.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py new file mode 100644 index 000000000..30be2eebc --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py @@ -0,0 +1,1152 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(32, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + cat = torch.cat([x_15, x2], dim=1) + x_15 = x2 = None + x_16 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + split_1 = x_24.split(64, dim=1) + x_24 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_25 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.silu(x_26, inplace=True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_31 = x_30 + x1_1 + x_30 = x1_1 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = x_37 + x_31 + x_37 = x_31 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x_38 + x_44 = x_38 = None + cat_1 = torch.cat([x_45, x2_1], dim=1) + x_45 = x2_1 = None + x_46 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + split_2 = x_54.split(128, dim=1) + x_54 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_55 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_56 = torch.nn.functional.batch_norm( + x_55, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_55 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_57 = torch.nn.functional.silu(x_56, inplace=True) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_59 = torch.nn.functional.batch_norm( + x_58, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_58 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_60 = torch.nn.functional.silu(x_59, inplace=True) + x_59 = None + x_61 = x_60 + x1_2 + x_60 = x1_2 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_64 = torch.nn.functional.silu(x_63, inplace=True) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_61 + x_67 = x_61 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = x_74 + x_68 + x_74 = x_68 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.silu(x_77, inplace=True) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = x_81 + x_75 + x_81 = x_75 = None + cat_2 = torch.cat([x_82, x2_2], dim=1) + x_82 = x2_2 = None + x_83 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + x_83, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_85 = torch.nn.functional.silu(x_84, inplace=True) + x_84 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + split_3 = x_91.split(256, dim=1) + x_91 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_92 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.silu(x_93, inplace=True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.silu(x_96, inplace=True) + x_96 = None + x_98 = x_97 + x1_3 + x_97 = x1_3 = None + cat_3 = torch.cat([x_98, x2_3], dim=1) + x_98 = x2_3 = None + x_99 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.silu(x_100, inplace=True) + x_100 = None + x_102 = torch.nn.functional.adaptive_avg_pool2d(x_101, 1) + x_101 = None + x_103 = x_102.flatten(1, -1) + x_102 = None + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_104 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_105,) diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py new file mode 100644 index 000000000..f525f7fe6 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py @@ -0,0 +1,1612 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.009 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.243 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.127 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..bf730a109 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +c3068f037595d3dc7ca29f97628506445b72363bc464669645887ead284dff75 \ No newline at end of file diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_l.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/model.py b/samples/timm/cs3darknet_l.c2ns_in1k/model.py new file mode 100644 index 000000000..5dd3a1f13 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/model.py @@ -0,0 +1,2041 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(64, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(128, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + cat_1 = torch.cat([x_83, x2_1], dim=1) + x_83 = x2_1 = None + x_84 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + split_2 = x_92.split(256, dim=1) + x_92 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_93 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = x_98 + x1_2 + x_98 = x1_2 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x_99 + x_105 = x_99 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + cat_2 = torch.cat([x_155, x2_2], dim=1) + x_155 = x2_2 = None + x_156 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + split_3 = x_164.split(512, dim=1) + x_164 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_165 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.silu(x_166, inplace=True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.silu(x_169, inplace=True) + x_169 = None + x_171 = x_170 + x1_3 + x_170 = x1_3 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.silu(x_173, inplace=True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.silu(x_176, inplace=True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.conv2d( + x_178, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_180 = torch.nn.functional.batch_norm( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_181 = torch.nn.functional.silu(x_180, inplace=True) + x_180 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_185 = x_184 + x_178 + x_184 = x_178 = None + cat_3 = torch.cat([x_185, x2_3], dim=1) + x_185 = x2_3 = None + x_186 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.nn.functional.adaptive_avg_pool2d(x_188, 1) + x_188 = None + x_190 = x_189.flatten(1, -1) + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_191 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_192,) diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..a4a7a3daa --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,2862 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.060 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..8df286919 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +cebfb5f58a4ad3894476513d82b2d34bd5c612a4662c6224d95dfb054105f459 \ No newline at end of file diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_m.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_m.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/model.py b/samples/timm/cs3darknet_m.c2ns_in1k/model.py new file mode 100644 index 000000000..0a058a1a8 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/model.py @@ -0,0 +1,1551 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(48, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + cat = torch.cat([x_25, x2], dim=1) + x_25 = x2 = None + x_26 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + split_1 = x_34.split(96, dim=1) + x_34 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_35 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = torch.conv2d( + x_37, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_37 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.silu(x_39, inplace=True) + x_39 = None + x_41 = x_40 + x1_1 + x_40 = x1_1 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x_41 + x_47 = x_41 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + cat_1 = torch.cat([x_62, x2_1], dim=1) + x_62 = x2_1 = None + x_63 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_68 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + split_2 = x_71.split(192, dim=1) + x_71 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_72 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.silu(x_76, inplace=True) + x_76 = None + x_78 = x_77 + x1_2 + x_77 = x1_2 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.silu(x_83, inplace=True) + x_83 = None + x_85 = x_84 + x_78 + x_84 = x_78 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_85 + x_91 = x_85 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = x_98 + x_92 + x_98 = x_92 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x_99 + x_105 = x_99 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + cat_2 = torch.cat([x_113, x2_2], dim=1) + x_113 = x2_2 = None + x_114 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_122 = torch.nn.functional.silu(x_121, inplace=True) + x_121 = None + split_3 = x_122.split(384, dim=1) + x_122 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_123 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.silu(x_124, inplace=True) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_129 = x_128 + x1_3 + x_128 = x1_3 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_135 = torch.nn.functional.silu(x_134, inplace=True) + x_134 = None + x_136 = x_135 + x_129 + x_135 = x_129 = None + cat_3 = torch.cat([x_136, x2_3], dim=1) + x_136 = x2_3 = None + x_137 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.silu(x_138, inplace=True) + x_138 = None + x_140 = torch.nn.functional.adaptive_avg_pool2d(x_139, 1) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..b5cf320a1 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py @@ -0,0 +1,2162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [24, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 24, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [96, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.141 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.203 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.207 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [768, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..8a4e63364 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +2d3ba3b73cded31585e47c5b7453f289cd329b85d1f959c316adaa2d4b0cbc92 \ No newline at end of file diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_x.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/model.py b/samples/timm/cs3darknet_x.c2ns_in1k/model.py new file mode 100644 index 000000000..4b502d175 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/model.py @@ -0,0 +1,2251 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(160, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = x_89 + x_83 + x_89 = x_83 = None + cat_1 = torch.cat([x_90, x2_1], dim=1) + x_90 = x2_1 = None + x_91 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + x_91, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_91 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_93 = torch.nn.functional.silu(x_92, inplace=True) + x_92 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_95 = torch.nn.functional.batch_norm( + x_94, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_94 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_96 = torch.nn.functional.silu(x_95, inplace=True) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + split_2 = x_99.split(320, dim=1) + x_99 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_100 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x1_2 + x_105 = x1_2 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = x_161 + x_155 + x_161 = x_155 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_164 = torch.nn.functional.batch_norm( + x_163, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_165 = torch.nn.functional.silu(x_164, inplace=True) + x_164 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_169 = x_168 + x_162 + x_168 = x_162 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = torch.conv2d( + x_172, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_172 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_174 = torch.nn.functional.batch_norm( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_173 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_175 = torch.nn.functional.silu(x_174, inplace=True) + x_174 = None + x_176 = x_175 + x_169 + x_175 = x_169 = None + cat_2 = torch.cat([x_176, x2_2], dim=1) + x_176 = x2_2 = None + x_177 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + split_3 = x_185.split(640, dim=1) + x_185 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_186 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + x_189, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_191 = torch.nn.functional.silu(x_190, inplace=True) + x_190 = None + x_192 = x_191 + x1_3 + x_191 = x1_3 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.silu(x_194, inplace=True) + x_194 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_197 = torch.nn.functional.batch_norm( + x_196, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_196 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_198 = torch.nn.functional.silu(x_197, inplace=True) + x_197 = None + x_199 = x_198 + x_192 + x_198 = x_192 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.silu(x_201, inplace=True) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_199 + x_205 = x_199 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..c0621a060 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,3162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.231 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.160 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt b/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt new file mode 100644 index 000000000..d019171a0 --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt @@ -0,0 +1 @@ +7a5a8bd289abb7ffad2c153c67f5dfa9567d4efcf89bf2788d22265ba9030dcd \ No newline at end of file diff --git a/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json b/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3edgenet_x.c2_in1k/input_meta.py b/samples/timm/cs3edgenet_x.c2_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3edgenet_x.c2_in1k/input_tensor_constraints.py b/samples/timm/cs3edgenet_x.c2_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3edgenet_x.c2_in1k/model.py b/samples/timm/cs3edgenet_x.c2_in1k/model.py new file mode 100644 index 000000000..611d6b788 --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/model.py @@ -0,0 +1,2251 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(160, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = x_89 + x_83 + x_89 = x_83 = None + cat_1 = torch.cat([x_90, x2_1], dim=1) + x_90 = x2_1 = None + x_91 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + x_91, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_91 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_93 = torch.nn.functional.silu(x_92, inplace=True) + x_92 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_95 = torch.nn.functional.batch_norm( + x_94, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_94 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_96 = torch.nn.functional.silu(x_95, inplace=True) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + split_2 = x_99.split(320, dim=1) + x_99 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_100 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x1_2 + x_105 = x1_2 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = x_161 + x_155 + x_161 = x_155 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_164 = torch.nn.functional.batch_norm( + x_163, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_165 = torch.nn.functional.silu(x_164, inplace=True) + x_164 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_169 = x_168 + x_162 + x_168 = x_162 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = torch.conv2d( + x_172, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_172 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_174 = torch.nn.functional.batch_norm( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_173 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_175 = torch.nn.functional.silu(x_174, inplace=True) + x_174 = None + x_176 = x_175 + x_169 + x_175 = x_169 = None + cat_2 = torch.cat([x_176, x2_2], dim=1) + x_176 = x2_2 = None + x_177 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + split_3 = x_185.split(640, dim=1) + x_185 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_186 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + x_189, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_191 = torch.nn.functional.silu(x_190, inplace=True) + x_190 = None + x_192 = x_191 + x1_3 + x_191 = x1_3 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.silu(x_194, inplace=True) + x_194 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_197 = torch.nn.functional.batch_norm( + x_196, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_196 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_198 = torch.nn.functional.silu(x_197, inplace=True) + x_197 = None + x_199 = x_198 + x_192 + x_198 = x_192 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.silu(x_201, inplace=True) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_199 + x_205 = x_199 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py b/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py new file mode 100644 index 000000000..a2264641a --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py @@ -0,0 +1,3162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..89a760bcf --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +72fd84a46be8697b4d0df714a13ffa325c87f75dd13f8b9b1fa1e3f20421a218 \ No newline at end of file diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py new file mode 100644 index 000000000..4482575d5 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(160, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_se_36 = x_95.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_96 = x_95 * sigmoid_9 + x_95 = sigmoid_9 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = x_99 + x_92 + x_99 = x_92 = None + cat_1 = torch.cat([x_100, x2_1], dim=1) + x_100 = x2_1 = None + x_101 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_102 = torch.nn.functional.batch_norm( + x_101, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_101 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_103 = torch.nn.functional.silu(x_102, inplace=True) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + split_2 = x_109.split(320, dim=1) + x_109 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_110 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x1_2 + x_116 = x1_2 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_se_72 = x_176.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_177 = x_176 * sigmoid_18 + x_176 = sigmoid_18 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.silu(x_179, inplace=True) + x_179 = None + x_181 = x_180 + x_173 + x_180 = x_173 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_se_76 = x_184.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_185 = x_184 * sigmoid_19 + x_184 = sigmoid_19 = None + x_186 = torch.conv2d( + x_185, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = x_188 + x_181 + x_188 = x_181 = None + x_190 = torch.conv2d( + x_189, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.silu(x_191, inplace=True) + x_191 = None + x_se_80 = x_192.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_193 = x_192 * sigmoid_20 + x_192 = sigmoid_20 = None + x_194 = torch.conv2d( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + x_194, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_194 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_196 = torch.nn.functional.silu(x_195, inplace=True) + x_195 = None + x_197 = x_196 + x_189 + x_196 = x_189 = None + cat_2 = torch.cat([x_197, x2_2], dim=1) + x_197 = x2_2 = None + x_198 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + x_198, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_198 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_200 = torch.nn.functional.silu(x_199, inplace=True) + x_199 = None + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + x_201, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_201 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_203 = torch.nn.functional.silu(x_202, inplace=True) + x_202 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_206 = torch.nn.functional.silu(x_205, inplace=True) + x_205 = None + split_3 = x_206.split(640, dim=1) + x_206 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_207 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_se_84 = x_209.mean((2, 3), keepdim=True) + x_se_85 = torch.conv2d( + x_se_84, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_84 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_86 = torch.nn.functional.silu(x_se_85, inplace=True) + x_se_85 = None + x_se_87 = torch.conv2d( + x_se_86, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_21 = x_se_87.sigmoid() + x_se_87 = None + x_210 = x_209 * sigmoid_21 + x_209 = sigmoid_21 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_210 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.silu(x_212, inplace=True) + x_212 = None + x_214 = x_213 + x1_3 + x_213 = x1_3 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_216 = torch.nn.functional.batch_norm( + x_215, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_215 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_217 = torch.nn.functional.silu(x_216, inplace=True) + x_216 = None + x_se_88 = x_217.mean((2, 3), keepdim=True) + x_se_89 = torch.conv2d( + x_se_88, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_88 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_90 = torch.nn.functional.silu(x_se_89, inplace=True) + x_se_89 = None + x_se_91 = torch.conv2d( + x_se_90, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_90 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_22 = x_se_91.sigmoid() + x_se_91 = None + x_218 = x_217 * sigmoid_22 + x_217 = sigmoid_22 = None + x_219 = torch.conv2d( + x_218, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_218 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + x_219, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_221 = torch.nn.functional.silu(x_220, inplace=True) + x_220 = None + x_222 = x_221 + x_214 + x_221 = x_214 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_224 = torch.nn.functional.batch_norm( + x_223, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_223 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_225 = torch.nn.functional.silu(x_224, inplace=True) + x_224 = None + x_se_92 = x_225.mean((2, 3), keepdim=True) + x_se_93 = torch.conv2d( + x_se_92, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_94 = torch.nn.functional.silu(x_se_93, inplace=True) + x_se_93 = None + x_se_95 = torch.conv2d( + x_se_94, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_94 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_23 = x_se_95.sigmoid() + x_se_95 = None + x_226 = x_225 * sigmoid_23 + x_225 = sigmoid_23 = None + x_227 = torch.conv2d( + x_226, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_226 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_229 = torch.nn.functional.silu(x_228, inplace=True) + x_228 = None + x_230 = x_229 + x_222 + x_229 = x_222 = None + cat_3 = torch.cat([x_230, x2_3], dim=1) + x_230 = x2_3 = None + x_231 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + x_231, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_231 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_233 = torch.nn.functional.silu(x_232, inplace=True) + x_232 = None + x_234 = torch.nn.functional.adaptive_avg_pool2d(x_233, 1) + x_233 = None + x_235 = x_234.flatten(1, -1) + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_236 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_237,) diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..a0f422010 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,4122 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.221 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.245 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.132 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.248 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.245 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.128 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.128 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..4fd012443 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +9968dcd4e6249cddec435975f4335d38d057d1e3be27cb9ab2f6f4cb2dbc677d \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/input_meta.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py new file mode 100644 index 000000000..b3d853722 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py @@ -0,0 +1,2776 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(64, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(128, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + cat_1 = torch.cat([x_92, x2_1], dim=1) + x_92 = x2_1 = None + x_93 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_101 = torch.nn.functional.silu(x_100, inplace=True) + x_100 = None + split_2 = x_101.split(256, dim=1) + x_101 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_102 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.silu(x_103, inplace=True) + x_103 = None + x_se_36 = x_104.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_105 = x_104 * sigmoid_9 + x_104 = sigmoid_9 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.silu(x_107, inplace=True) + x_107 = None + x_109 = x_108 + x1_2 + x_108 = x1_2 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x_109 + x_116 = x_109 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + cat_2 = torch.cat([x_173, x2_2], dim=1) + x_173 = x2_2 = None + x_174 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_177 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + split_3 = x_182.split(512, dim=1) + x_182 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_183 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + x_se_72 = x_185.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_186 = x_185 * sigmoid_18 + x_185 = sigmoid_18 = None + x_187 = torch.conv2d( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_187 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_189 = torch.nn.functional.silu(x_188, inplace=True) + x_188 = None + x_190 = x_189 + x1_3 + x_189 = x1_3 = None + x_191 = torch.conv2d( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_192 = torch.nn.functional.batch_norm( + x_191, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_191 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_193 = torch.nn.functional.silu(x_192, inplace=True) + x_192 = None + x_se_76 = x_193.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_194 = x_193 * sigmoid_19 + x_193 = sigmoid_19 = None + x_195 = torch.conv2d( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_197 = torch.nn.functional.silu(x_196, inplace=True) + x_196 = None + x_198 = x_197 + x_190 + x_197 = x_190 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_200 = torch.nn.functional.batch_norm( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_199 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_201 = torch.nn.functional.silu(x_200, inplace=True) + x_200 = None + x_se_80 = x_201.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_202 = x_201 * sigmoid_20 + x_201 = sigmoid_20 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_198 + x_205 = x_198 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..972738598 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,3702 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.363 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.180 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.174 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.016 + std = 0.358 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.173 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.012 + std = 0.373 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.242 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.254 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.252 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.122 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.247 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.249 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.250 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..18bda7171 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +1b92ad067f29b181b3dc7b6bff02d3202367a7fe6d248d63413fad08d6f86697 \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py new file mode 100644 index 000000000..8091febe4 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(160, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_se_36 = x_95.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_96 = x_95 * sigmoid_9 + x_95 = sigmoid_9 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = x_99 + x_92 + x_99 = x_92 = None + cat_1 = torch.cat([x_100, x2_1], dim=1) + x_100 = x2_1 = None + x_101 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_102 = torch.nn.functional.batch_norm( + x_101, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_101 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_103 = torch.nn.functional.silu(x_102, inplace=True) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + split_2 = x_109.split(320, dim=1) + x_109 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_110 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x1_2 + x_116 = x1_2 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_se_72 = x_176.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_177 = x_176 * sigmoid_18 + x_176 = sigmoid_18 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.silu(x_179, inplace=True) + x_179 = None + x_181 = x_180 + x_173 + x_180 = x_173 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_se_76 = x_184.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_185 = x_184 * sigmoid_19 + x_184 = sigmoid_19 = None + x_186 = torch.conv2d( + x_185, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = x_188 + x_181 + x_188 = x_181 = None + x_190 = torch.conv2d( + x_189, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.silu(x_191, inplace=True) + x_191 = None + x_se_80 = x_192.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_193 = x_192 * sigmoid_20 + x_192 = sigmoid_20 = None + x_194 = torch.conv2d( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + x_194, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_194 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_196 = torch.nn.functional.silu(x_195, inplace=True) + x_195 = None + x_197 = x_196 + x_189 + x_196 = x_189 = None + cat_2 = torch.cat([x_197, x2_2], dim=1) + x_197 = x2_2 = None + x_198 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + x_198, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_198 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_200 = torch.nn.functional.silu(x_199, inplace=True) + x_199 = None + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + x_201, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_201 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_203 = torch.nn.functional.silu(x_202, inplace=True) + x_202 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_206 = torch.nn.functional.silu(x_205, inplace=True) + x_205 = None + split_3 = x_206.split(640, dim=1) + x_206 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_207 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_se_84 = x_209.mean((2, 3), keepdim=True) + x_se_85 = torch.conv2d( + x_se_84, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_84 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_86 = torch.nn.functional.silu(x_se_85, inplace=True) + x_se_85 = None + x_se_87 = torch.conv2d( + x_se_86, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_21 = x_se_87.sigmoid() + x_se_87 = None + x_210 = x_209 * sigmoid_21 + x_209 = sigmoid_21 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_210 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.silu(x_212, inplace=True) + x_212 = None + x_214 = x_213 + x1_3 + x_213 = x1_3 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_216 = torch.nn.functional.batch_norm( + x_215, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_215 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_217 = torch.nn.functional.silu(x_216, inplace=True) + x_216 = None + x_se_88 = x_217.mean((2, 3), keepdim=True) + x_se_89 = torch.conv2d( + x_se_88, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_88 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_90 = torch.nn.functional.silu(x_se_89, inplace=True) + x_se_89 = None + x_se_91 = torch.conv2d( + x_se_90, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_90 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_22 = x_se_91.sigmoid() + x_se_91 = None + x_218 = x_217 * sigmoid_22 + x_217 = sigmoid_22 = None + x_219 = torch.conv2d( + x_218, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_218 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + x_219, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_221 = torch.nn.functional.silu(x_220, inplace=True) + x_220 = None + x_222 = x_221 + x_214 + x_221 = x_214 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_224 = torch.nn.functional.batch_norm( + x_223, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_223 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_225 = torch.nn.functional.silu(x_224, inplace=True) + x_224 = None + x_se_92 = x_225.mean((2, 3), keepdim=True) + x_se_93 = torch.conv2d( + x_se_92, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_94 = torch.nn.functional.silu(x_se_93, inplace=True) + x_se_93 = None + x_se_95 = torch.conv2d( + x_se_94, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_94 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_23 = x_se_95.sigmoid() + x_se_95 = None + x_226 = x_225 * sigmoid_23 + x_225 = sigmoid_23 = None + x_227 = torch.conv2d( + x_226, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_226 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_229 = torch.nn.functional.silu(x_228, inplace=True) + x_228 = None + x_230 = x_229 + x_222 + x_229 = x_222 = None + cat_3 = torch.cat([x_230, x2_3], dim=1) + x_230 = x2_3 = None + x_231 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + x_231, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_231 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_233 = torch.nn.functional.silu(x_232, inplace=True) + x_232 = None + x_234 = torch.nn.functional.adaptive_avg_pool2d(x_233, 1) + x_233 = None + x_235 = x_234.flatten(1, -1) + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_236 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_237,) diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..4a7c009e8 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,4122 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.221 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.505 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.150 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.156 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.020 + std = 0.498 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.155 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.020 + std = 0.503 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.154 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.029 + std = 0.484 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.013 + std = 0.499 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.114 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.486 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.503 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.109 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.491 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.114 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.018 + std = 0.501 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.484 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.110 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.078 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.292 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.290 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.290 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.291 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.292 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.223 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.224 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.225 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt b/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt new file mode 100644 index 000000000..1a7ee185d --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +4510f5a457c81963c81e8519198a97e4628f0f5671dd4fe01bce02082834404a \ No newline at end of file diff --git a/samples/timm/cspdarknet53.ra_in1k/graph_net.json b/samples/timm/cspdarknet53.ra_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspdarknet53.ra_in1k/input_meta.py b/samples/timm/cspdarknet53.ra_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspdarknet53.ra_in1k/input_tensor_constraints.py b/samples/timm/cspdarknet53.ra_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspdarknet53.ra_in1k/model.py b/samples/timm/cspdarknet53.ra_in1k/model.py new file mode 100644 index 000000000..28566124b --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/model.py @@ -0,0 +1,2417 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + split = x_8.split(64, dim=1) + x_8 = None + xs = split[0] + xb = split[1] + split = None + x_9 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, True) + x_13 = None + x_15 = x_14 + xb + x_14 = xb = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + xb_1 = x_18.contiguous() + x_18 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_19 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, True) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_24 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + split_1 = x_27.split(64, dim=1) + x_27 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_28 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_33 = torch.nn.functional.leaky_relu(x_32, 0.01, True) + x_32 = None + x_34 = x_33 + xb_2 + x_33 = xb_2 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + x_38 = torch.conv2d( + x_37, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_37 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = x_40 + x_34 + x_40 = x_34 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.leaky_relu(x_43, 0.01, True) + x_43 = None + xb_3 = x_44.contiguous() + x_44 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_45 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.leaky_relu(x_46, 0.01, True) + x_46 = None + x_48 = torch.conv2d( + x_47, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_49 = torch.nn.functional.batch_norm( + x_48, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_48 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_50 = torch.nn.functional.leaky_relu(x_49, 0.01, True) + x_49 = None + x_51 = torch.conv2d( + x_50, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_50 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_52 = torch.nn.functional.batch_norm( + x_51, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_51 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_53 = torch.nn.functional.leaky_relu(x_52, 0.01, True) + x_52 = None + split_2 = x_53.split(128, dim=1) + x_53 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_54 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + xb_4 + x_59 = xb_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = x_94 + x_88 + x_94 = x_88 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.leaky_relu(x_97, 0.01, True) + x_97 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = x_101 + x_95 + x_101 = x_95 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.leaky_relu(x_104, 0.01, True) + x_104 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = x_108 + x_102 + x_108 = x_102 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + xb_5 = x_112.contiguous() + x_112 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_113 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_118 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + x_119, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_119 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_121 = torch.nn.functional.leaky_relu(x_120, 0.01, True) + x_120 = None + split_3 = x_121.split(256, dim=1) + x_121 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_122 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.leaky_relu(x_123, 0.01, True) + x_123 = None + x_125 = torch.conv2d( + x_124, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_124 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_126 = torch.nn.functional.batch_norm( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_127 = torch.nn.functional.leaky_relu(x_126, 0.01, True) + x_126 = None + x_128 = x_127 + xb_6 + x_127 = xb_6 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_130 = torch.nn.functional.batch_norm( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_131 = torch.nn.functional.leaky_relu(x_130, 0.01, True) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_133 = torch.nn.functional.batch_norm( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_132 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_134 = torch.nn.functional.leaky_relu(x_133, 0.01, True) + x_133 = None + x_135 = x_134 + x_128 + x_134 = x_128 = None + x_136 = torch.conv2d( + x_135, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_137 = torch.nn.functional.batch_norm( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_138 = torch.nn.functional.leaky_relu(x_137, 0.01, True) + x_137 = None + x_139 = torch.conv2d( + x_138, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_138 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_141 = torch.nn.functional.leaky_relu(x_140, 0.01, True) + x_140 = None + x_142 = x_141 + x_135 + x_141 = x_135 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.leaky_relu(x_147, 0.01, True) + x_147 = None + x_149 = x_148 + x_142 + x_148 = x_142 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.leaky_relu(x_151, 0.01, True) + x_151 = None + x_153 = torch.conv2d( + x_152, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_152 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_154 = torch.nn.functional.batch_norm( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_155 = torch.nn.functional.leaky_relu(x_154, 0.01, True) + x_154 = None + x_156 = x_155 + x_149 + x_155 = x_149 = None + x_157 = torch.conv2d( + x_156, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_158 = torch.nn.functional.batch_norm( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_157 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_159 = torch.nn.functional.leaky_relu(x_158, 0.01, True) + x_158 = None + x_160 = torch.conv2d( + x_159, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_159 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_161 = torch.nn.functional.batch_norm( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_162 = torch.nn.functional.leaky_relu(x_161, 0.01, True) + x_161 = None + x_163 = x_162 + x_156 + x_162 = x_156 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_166 = torch.nn.functional.leaky_relu(x_165, 0.01, True) + x_165 = None + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + x_167, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_167 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_169 = torch.nn.functional.leaky_relu(x_168, 0.01, True) + x_168 = None + x_170 = x_169 + x_163 + x_169 = x_163 = None + x_171 = torch.conv2d( + x_170, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.leaky_relu(x_175, 0.01, True) + x_175 = None + x_177 = x_176 + x_170 + x_176 = x_170 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.leaky_relu(x_179, 0.01, True) + x_179 = None + xb_7 = x_180.contiguous() + x_180 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_181 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_182 = torch.nn.functional.batch_norm( + x_181, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_181 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_183 = torch.nn.functional.leaky_relu(x_182, 0.01, True) + x_182 = None + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_185 = torch.nn.functional.batch_norm( + x_184, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_184 = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_186 = torch.nn.functional.leaky_relu(x_185, 0.01, True) + x_185 = None + x_187 = torch.conv2d( + x_186, + l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_186 = l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + x_187, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_187 = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_189 = torch.nn.functional.leaky_relu(x_188, 0.01, True) + x_188 = None + split_4 = x_189.split(512, dim=1) + x_189 = None + xs_4 = split_4[0] + xb_8 = split_4[1] + split_4 = None + x_190 = torch.conv2d( + xb_8, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.leaky_relu(x_191, 0.01, True) + x_191 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_192 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.leaky_relu(x_194, 0.01, True) + x_194 = None + x_196 = x_195 + xb_8 + x_195 = xb_8 = None + x_197 = torch.conv2d( + x_196, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_198 = torch.nn.functional.batch_norm( + x_197, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_197 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_199 = torch.nn.functional.leaky_relu(x_198, 0.01, True) + x_198 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_199 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.leaky_relu(x_201, 0.01, True) + x_201 = None + x_203 = x_202 + x_196 + x_202 = x_196 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_206 = torch.nn.functional.leaky_relu(x_205, 0.01, True) + x_205 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_206 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.leaky_relu(x_208, 0.01, True) + x_208 = None + x_210 = x_209 + x_203 + x_209 = x_203 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.leaky_relu(x_212, 0.01, True) + x_212 = None + x_214 = torch.conv2d( + x_213, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_215 = torch.nn.functional.batch_norm( + x_214, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_214 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_216 = torch.nn.functional.leaky_relu(x_215, 0.01, True) + x_215 = None + x_217 = x_216 + x_210 + x_216 = x_210 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_219 = torch.nn.functional.batch_norm( + x_218, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_218 = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_220 = torch.nn.functional.leaky_relu(x_219, 0.01, True) + x_219 = None + xb_9 = x_220.contiguous() + x_220 = None + cat_4 = torch.cat([xs_4, xb_9], dim=1) + xs_4 = xb_9 = None + x_221 = torch.conv2d( + cat_4, + l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_4 = l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + x_221, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_221 = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_223 = torch.nn.functional.leaky_relu(x_222, 0.01, True) + x_222 = None + x_224 = torch.nn.functional.adaptive_avg_pool2d(x_223, 1) + x_223 = None + x_225 = x_224.flatten(1, -1) + x_224 = None + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_226 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_227,) diff --git a/samples/timm/cspdarknet53.ra_in1k/weight_meta.py b/samples/timm/cspdarknet53.ra_in1k/weight_meta.py new file mode 100644 index 000000000..4a1ed3882 --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/weight_meta.py @@ -0,0 +1,3418 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.252 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspresnet50.ra_in1k/graph_hash.txt b/samples/timm/cspresnet50.ra_in1k/graph_hash.txt new file mode 100644 index 000000000..5d28f39ff --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +d7eb07ed68b44726e3fb9f25d944a7607852e83651394761699a126e70a4755e \ No newline at end of file diff --git a/samples/timm/cspresnet50.ra_in1k/graph_net.json b/samples/timm/cspresnet50.ra_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspresnet50.ra_in1k/input_meta.py b/samples/timm/cspresnet50.ra_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspresnet50.ra_in1k/input_tensor_constraints.py b/samples/timm/cspresnet50.ra_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspresnet50.ra_in1k/model.py b/samples/timm/cspresnet50.ra_in1k/model.py new file mode 100644 index 000000000..34bd70bbe --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/model.py @@ -0,0 +1,1964 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch.nn.functional.max_pool2d( + x_2, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split = x_4.split(128, dim=1) + x_4 = None + xs = split[0] + xb = split[1] + split = None + x_5 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_6 = torch.nn.functional.batch_norm( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_7 = torch.nn.functional.leaky_relu(x_6, 0.01, True) + x_6 = None + x_8 = torch.conv2d( + x_7, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_7 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_10 = torch.nn.functional.leaky_relu(x_9, 0.01, True) + x_9 = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_13 = x_12 + xb + x_12 = xb = None + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, False) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.leaky_relu(x_16, 0.01, True) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_19 = torch.nn.functional.batch_norm( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_20 = torch.nn.functional.leaky_relu(x_19, 0.01, True) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_22 = torch.nn.functional.batch_norm( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_23 = x_22 + x_14 + x_22 = x_14 = None + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, False) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_33 = x_32 + x_24 + x_32 = x_24 = None + x_34 = torch.nn.functional.leaky_relu(x_33, 0.01, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + xb_1 = x_37.contiguous() + x_37 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_38 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + x_41, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_43 = torch.nn.functional.leaky_relu(x_42, 0.01, True) + x_42 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_45 = torch.nn.functional.batch_norm( + x_44, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_44 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_1 = x_45.split(256, dim=1) + x_45 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_46 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.leaky_relu(x_47, 0.01, True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.leaky_relu(x_50, 0.01, True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_54 = x_53 + xb_2 + x_53 = xb_2 = None + x_55 = torch.nn.functional.leaky_relu(x_54, 0.01, False) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.leaky_relu(x_57, 0.01, True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.leaky_relu(x_60, 0.01, True) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_64 = x_63 + x_55 + x_63 = x_55 = None + x_65 = torch.nn.functional.leaky_relu(x_64, 0.01, False) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.leaky_relu(x_67, 0.01, True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_68 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.leaky_relu(x_70, 0.01, True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_74 = x_73 + x_65 + x_73 = x_65 = None + x_75 = torch.nn.functional.leaky_relu(x_74, 0.01, False) + x_74 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_75 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.leaky_relu(x_77, 0.01, True) + x_77 = None + xb_3 = x_78.contiguous() + x_78 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_79 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.leaky_relu(x_80, 0.01, True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_2 = x_86.split(512, dim=1) + x_86 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_87 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.leaky_relu(x_88, 0.01, True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.leaky_relu(x_91, 0.01, True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_95 = x_94 + xb_4 + x_94 = xb_4 = None + x_96 = torch.nn.functional.leaky_relu(x_95, 0.01, False) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.leaky_relu(x_98, 0.01, True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.leaky_relu(x_101, 0.01, True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_105 = x_104 + x_96 + x_104 = x_96 = None + x_106 = torch.nn.functional.leaky_relu(x_105, 0.01, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.leaky_relu(x_108, 0.01, True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_115 = x_114 + x_106 + x_114 = x_106 = None + x_116 = torch.nn.functional.leaky_relu(x_115, 0.01, False) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.leaky_relu(x_118, 0.01, True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = (None) + x_125 = x_124 + x_116 + x_124 = x_116 = None + x_126 = torch.nn.functional.leaky_relu(x_125, 0.01, False) + x_125 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = (None) + x_135 = x_134 + x_126 + x_134 = x_126 = None + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + xb_5 = x_139.contiguous() + x_139 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_140 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + x_140, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_140 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_142 = torch.nn.functional.leaky_relu(x_141, 0.01, True) + x_141 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_142 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_3 = x_147.split(1024, dim=1) + x_147 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_148 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_156 = x_155 + xb_6 + x_155 = xb_6 = None + x_157 = torch.nn.functional.leaky_relu(x_156, 0.01, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_166 = x_165 + x_157 + x_165 = x_157 = None + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, False) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + xb_7 = x_170.contiguous() + x_170 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_171 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.nn.functional.adaptive_avg_pool2d(x_173, 1) + x_173 = None + x_175 = x_174.flatten(1, -1) + x_174 = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_176 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_177,) diff --git a/samples/timm/cspresnet50.ra_in1k/weight_meta.py b/samples/timm/cspresnet50.ra_in1k/weight_meta.py new file mode 100644 index 000000000..48413f25f --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/weight_meta.py @@ -0,0 +1,2810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspresnext50.ra_in1k/graph_hash.txt b/samples/timm/cspresnext50.ra_in1k/graph_hash.txt new file mode 100644 index 000000000..99b0ca8e4 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +828c24b6529b054ebd739d9b467e9b368a14d6bd6582a8549964e3483aee93df \ No newline at end of file diff --git a/samples/timm/cspresnext50.ra_in1k/graph_net.json b/samples/timm/cspresnext50.ra_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspresnext50.ra_in1k/input_meta.py b/samples/timm/cspresnext50.ra_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspresnext50.ra_in1k/input_tensor_constraints.py b/samples/timm/cspresnext50.ra_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/cspresnext50.ra_in1k/model.py b/samples/timm/cspresnext50.ra_in1k/model.py new file mode 100644 index 000000000..6ca320a88 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/model.py @@ -0,0 +1,1964 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch.nn.functional.max_pool2d( + x_2, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split = x_4.split(128, dim=1) + x_4 = None + xs = split[0] + xb = split[1] + split = None + x_5 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_6 = torch.nn.functional.batch_norm( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_7 = torch.nn.functional.leaky_relu(x_6, 0.01, True) + x_6 = None + x_8 = torch.conv2d( + x_7, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_7 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_10 = torch.nn.functional.leaky_relu(x_9, 0.01, True) + x_9 = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_13 = x_12 + xb + x_12 = xb = None + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, False) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.leaky_relu(x_16, 0.01, True) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_19 = torch.nn.functional.batch_norm( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_20 = torch.nn.functional.leaky_relu(x_19, 0.01, True) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_22 = torch.nn.functional.batch_norm( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_23 = x_22 + x_14 + x_22 = x_14 = None + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, False) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_33 = x_32 + x_24 + x_32 = x_24 = None + x_34 = torch.nn.functional.leaky_relu(x_33, 0.01, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + xb_1 = x_37.contiguous() + x_37 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_38 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_40 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + x_41, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_43 = torch.nn.functional.leaky_relu(x_42, 0.01, True) + x_42 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_45 = torch.nn.functional.batch_norm( + x_44, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_44 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_1 = x_45.split(256, dim=1) + x_45 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_46 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.leaky_relu(x_47, 0.01, True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.leaky_relu(x_50, 0.01, True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_54 = x_53 + xb_2 + x_53 = xb_2 = None + x_55 = torch.nn.functional.leaky_relu(x_54, 0.01, False) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.leaky_relu(x_57, 0.01, True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.leaky_relu(x_60, 0.01, True) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_64 = x_63 + x_55 + x_63 = x_55 = None + x_65 = torch.nn.functional.leaky_relu(x_64, 0.01, False) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.leaky_relu(x_67, 0.01, True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_68 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.leaky_relu(x_70, 0.01, True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_74 = x_73 + x_65 + x_73 = x_65 = None + x_75 = torch.nn.functional.leaky_relu(x_74, 0.01, False) + x_74 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_75 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.leaky_relu(x_77, 0.01, True) + x_77 = None + xb_3 = x_78.contiguous() + x_78 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_79 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.leaky_relu(x_80, 0.01, True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_81 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_2 = x_86.split(512, dim=1) + x_86 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_87 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.leaky_relu(x_88, 0.01, True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.leaky_relu(x_91, 0.01, True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_95 = x_94 + xb_4 + x_94 = xb_4 = None + x_96 = torch.nn.functional.leaky_relu(x_95, 0.01, False) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.leaky_relu(x_98, 0.01, True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.leaky_relu(x_101, 0.01, True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_105 = x_104 + x_96 + x_104 = x_96 = None + x_106 = torch.nn.functional.leaky_relu(x_105, 0.01, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.leaky_relu(x_108, 0.01, True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_115 = x_114 + x_106 + x_114 = x_106 = None + x_116 = torch.nn.functional.leaky_relu(x_115, 0.01, False) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.leaky_relu(x_118, 0.01, True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = (None) + x_125 = x_124 + x_116 + x_124 = x_116 = None + x_126 = torch.nn.functional.leaky_relu(x_125, 0.01, False) + x_125 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = (None) + x_135 = x_134 + x_126 + x_134 = x_126 = None + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + xb_5 = x_139.contiguous() + x_139 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_140 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + x_140, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_140 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_142 = torch.nn.functional.leaky_relu(x_141, 0.01, True) + x_141 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_142 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_3 = x_147.split(1024, dim=1) + x_147 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_148 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_156 = x_155 + xb_6 + x_155 = xb_6 = None + x_157 = torch.nn.functional.leaky_relu(x_156, 0.01, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_166 = x_165 + x_157 + x_165 = x_157 = None + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, False) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + xb_7 = x_170.contiguous() + x_170 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_171 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.nn.functional.adaptive_avg_pool2d(x_173, 1) + x_173 = None + x_175 = x_174.flatten(1, -1) + x_174 = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_176 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_177,) diff --git a/samples/timm/cspresnext50.ra_in1k/weight_meta.py b/samples/timm/cspresnext50.ra_in1k/weight_meta.py new file mode 100644 index 000000000..ffe972c49 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/weight_meta.py @@ -0,0 +1,2810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.087 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [2048, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [2048, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/darknet53.c2ns_in1k/graph_hash.txt b/samples/timm/darknet53.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..9cdb76c82 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +1c50cabf9a25ff8ea22ef875728acf4e6412d099b4dfaf536614bf27e8a11e73 \ No newline at end of file diff --git a/samples/timm/darknet53.c2ns_in1k/graph_net.json b/samples/timm/darknet53.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/darknet53.c2ns_in1k/input_meta.py b/samples/timm/darknet53.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/darknet53.c2ns_in1k/input_tensor_constraints.py b/samples/timm/darknet53.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/darknet53.c2ns_in1k/model.py b/samples/timm/darknet53.c2ns_in1k/model.py new file mode 100644 index 000000000..3245af82e --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/model.py @@ -0,0 +1,1857 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = x_11 + x_5 + x_11 = x_5 = None + x_13 = torch.conv2d( + x_12, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_12 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_14 = torch.nn.functional.batch_norm( + x_13, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_13 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_15 = torch.nn.functional.leaky_relu(x_14, 0.01, True) + x_14 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.leaky_relu(x_24, 0.01, True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.leaky_relu(x_27, 0.01, True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + x_30 = torch.conv2d( + x_29, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_29 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.leaky_relu(x_31, 0.01, True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.leaky_relu(x_34, 0.01, True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.leaky_relu(x_37, 0.01, True) + x_37 = None + x_39 = x_38 + x_32 + x_38 = x_32 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_41 = torch.nn.functional.batch_norm( + x_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_42 = torch.nn.functional.leaky_relu(x_41, 0.01, True) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_44 = torch.nn.functional.batch_norm( + x_43, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_43 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_45 = torch.nn.functional.leaky_relu(x_44, 0.01, True) + x_44 = None + x_46 = x_45 + x_39 + x_45 = x_39 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_48 = torch.nn.functional.batch_norm( + x_47, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_47 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_49 = torch.nn.functional.leaky_relu(x_48, 0.01, True) + x_48 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_49 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + x_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_52 = torch.nn.functional.leaky_relu(x_51, 0.01, True) + x_51 = None + x_53 = x_52 + x_46 + x_52 = x_46 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + x_53 + x_59 = x_53 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.leaky_relu(x_96, 0.01, True) + x_96 = None + x_98 = x_97 + x_91 + x_97 = x_91 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = torch.conv2d( + x_101, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.leaky_relu(x_103, 0.01, True) + x_103 = None + x_105 = x_104 + x_98 + x_104 = x_98 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = torch.conv2d( + x_108, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_108 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_110 = torch.nn.functional.batch_norm( + x_109, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_111 = torch.nn.functional.leaky_relu(x_110, 0.01, True) + x_110 = None + x_112 = x_111 + x_105 + x_111 = x_105 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = x_118 + x_112 + x_118 = x_112 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.leaky_relu(x_124, 0.01, True) + x_124 = None + x_126 = x_125 + x_119 + x_125 = x_119 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, True) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + x_140 = x_139 + x_133 + x_139 = x_133 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_142 = torch.nn.functional.batch_norm( + x_141, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_141 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_143 = torch.nn.functional.leaky_relu(x_142, 0.01, True) + x_142 = None + x_144 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + x_144, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_144 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_146 = torch.nn.functional.leaky_relu(x_145, 0.01, True) + x_145 = None + x_147 = x_146 + x_140 + x_146 = x_140 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.leaky_relu(x_155, 0.01, True) + x_155 = None + x_157 = x_156 + x_150 + x_156 = x_150 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = x_163 + x_157 + x_163 = x_157 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + x_171 = x_170 + x_164 + x_170 = x_164 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.leaky_relu(x_173, 0.01, True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.leaky_relu(x_176, 0.01, True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.nn.functional.adaptive_avg_pool2d(x_178, 1) + x_178 = None + x_180 = x_179.flatten(1, -1) + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_181 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_182,) diff --git a/samples/timm/darknet53.c2ns_in1k/weight_meta.py b/samples/timm/darknet53.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..10302cd69 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/weight_meta.py @@ -0,0 +1,2648 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.254 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt b/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt new file mode 100644 index 000000000..6449c61d9 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +c2ae90f0c65938bc0418ed7fc311e7f62365ebea543cf67c922c46d0878fab01 \ No newline at end of file diff --git a/samples/timm/darknetaa53.c2ns_in1k/graph_net.json b/samples/timm/darknetaa53.c2ns_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/darknetaa53.c2ns_in1k/input_meta.py b/samples/timm/darknetaa53.c2ns_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/darknetaa53.c2ns_in1k/input_tensor_constraints.py b/samples/timm/darknetaa53.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/darknetaa53.c2ns_in1k/model.py b/samples/timm/darknetaa53.c2ns_in1k/model.py new file mode 100644 index 000000000..67732f4a6 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/model.py @@ -0,0 +1,1867 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch._C._nn.avg_pool2d(x_2, 2, 2, 0, False, True, None) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = x_11 + x_5 + x_11 = x_5 = None + input_2 = torch._C._nn.avg_pool2d(x_12, 2, 2, 0, False, True, None) + x_12 = None + x_13 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_14 = torch.nn.functional.batch_norm( + x_13, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_13 = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_15 = torch.nn.functional.leaky_relu(x_14, 0.01, True) + x_14 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.leaky_relu(x_24, 0.01, True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.leaky_relu(x_27, 0.01, True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + input_3 = torch._C._nn.avg_pool2d(x_29, 2, 2, 0, False, True, None) + x_29 = None + x_30 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_3 = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.leaky_relu(x_31, 0.01, True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.leaky_relu(x_34, 0.01, True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.leaky_relu(x_37, 0.01, True) + x_37 = None + x_39 = x_38 + x_32 + x_38 = x_32 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_41 = torch.nn.functional.batch_norm( + x_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_42 = torch.nn.functional.leaky_relu(x_41, 0.01, True) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_44 = torch.nn.functional.batch_norm( + x_43, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_43 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_45 = torch.nn.functional.leaky_relu(x_44, 0.01, True) + x_44 = None + x_46 = x_45 + x_39 + x_45 = x_39 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_48 = torch.nn.functional.batch_norm( + x_47, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_47 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_49 = torch.nn.functional.leaky_relu(x_48, 0.01, True) + x_48 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_49 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + x_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_52 = torch.nn.functional.leaky_relu(x_51, 0.01, True) + x_51 = None + x_53 = x_52 + x_46 + x_52 = x_46 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + x_53 + x_59 = x_53 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + input_4 = torch._C._nn.avg_pool2d(x_88, 2, 2, 0, False, True, None) + x_88 = None + x_89 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_4 = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.leaky_relu(x_96, 0.01, True) + x_96 = None + x_98 = x_97 + x_91 + x_97 = x_91 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = torch.conv2d( + x_101, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.leaky_relu(x_103, 0.01, True) + x_103 = None + x_105 = x_104 + x_98 + x_104 = x_98 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = torch.conv2d( + x_108, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_108 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_110 = torch.nn.functional.batch_norm( + x_109, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_111 = torch.nn.functional.leaky_relu(x_110, 0.01, True) + x_110 = None + x_112 = x_111 + x_105 + x_111 = x_105 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = x_118 + x_112 + x_118 = x_112 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.leaky_relu(x_124, 0.01, True) + x_124 = None + x_126 = x_125 + x_119 + x_125 = x_119 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, True) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + x_140 = x_139 + x_133 + x_139 = x_133 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_142 = torch.nn.functional.batch_norm( + x_141, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_141 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_143 = torch.nn.functional.leaky_relu(x_142, 0.01, True) + x_142 = None + x_144 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + x_144, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_144 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_146 = torch.nn.functional.leaky_relu(x_145, 0.01, True) + x_145 = None + x_147 = x_146 + x_140 + x_146 = x_140 = None + input_5 = torch._C._nn.avg_pool2d(x_147, 2, 2, 0, False, True, None) + x_147 = None + x_148 = torch.conv2d( + input_5, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_5 = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.leaky_relu(x_155, 0.01, True) + x_155 = None + x_157 = x_156 + x_150 + x_156 = x_150 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = x_163 + x_157 + x_163 = x_157 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + x_171 = x_170 + x_164 + x_170 = x_164 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.leaky_relu(x_173, 0.01, True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.leaky_relu(x_176, 0.01, True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.nn.functional.adaptive_avg_pool2d(x_178, 1) + x_178 = None + x_180 = x_179.flatten(1, -1) + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_181 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_182,) diff --git a/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py b/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py new file mode 100644 index 000000000..9ef5d7cca --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py @@ -0,0 +1,2638 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.078 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.246 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..409c645a1 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +f6582d13c42b77e560d755b7c97489429c32219c0d78a2df64567083e29ff3e6 \ No newline at end of file diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_base_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_base_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/model.py b/samples/timm/deit3_base_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..4ecba32b8 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..7844b32df --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..ca69e6da2 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +707327cac3944d79a30f7b30bf5570beb58057638f3953b5b7d3a48b003a5750 \ No newline at end of file diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/input_meta.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py new file mode 100644 index 000000000..0edee0fb1 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py @@ -0,0 +1,4788 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (14, 14), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1280,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 257, 3, 16, 80) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 257, 1280) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (1280,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (1280,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 257, 3, 16, 80) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 257, 1280) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (1280,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (1280,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 257, 3, 16, 80) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 257, 1280) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (1280,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (1280,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 257, 3, 16, 80) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 257, 1280) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1280,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1280,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 257, 3, 16, 80) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 257, 1280) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (1280,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (1280,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 257, 3, 16, 80) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 257, 1280) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (1280,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1280,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 257, 3, 16, 80) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 257, 1280) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (1280,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (1280,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 257, 3, 16, 80) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 257, 1280) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1280,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (1280,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 257, 3, 16, 80) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 257, 1280) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (1280,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (1280,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 257, 3, 16, 80) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 257, 1280) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1280,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (1280,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 257, 3, 16, 80) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 257, 1280) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (1280,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (1280,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 257, 3, 16, 80) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 257, 1280) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1280,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (1280,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_161 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 257, 3, 16, 80) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_13.reshape(1, 257, 1280) + transpose_13 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + mul_24 = x_165 * l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + x_165 = l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = None + x_166 = x_160 + mul_24 + x_160 = mul_24 = None + x_167 = torch.nn.functional.layer_norm( + x_166, + (1280,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + x_167, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_167 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_25 = x_172 * l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + x_172 = l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = None + x_173 = x_166 + mul_25 + x_166 = mul_25 = None + x_174 = torch.nn.functional.layer_norm( + x_173, + (1280,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_174 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 257, 3, 16, 80) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_175 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_175.transpose(1, 2) + x_175 = None + x_176 = transpose_14.reshape(1, 257, 1280) + transpose_14 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_176 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_178 = torch.nn.functional.dropout(x_177, 0.0, False, False) + x_177 = None + mul_26 = x_178 * l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + x_178 = l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = None + x_179 = x_173 + mul_26 + x_173 = mul_26 = None + x_180 = torch.nn.functional.layer_norm( + x_179, + (1280,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_182 = torch._C._nn.gelu(x_181, approximate="none") + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + mul_27 = x_185 * l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + x_185 = l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = None + x_186 = x_179 + mul_27 + x_179 = mul_27 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1280,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_187, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_187 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 257, 3, 16, 80) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_188 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_188.transpose(1, 2) + x_188 = None + x_189 = transpose_15.reshape(1, 257, 1280) + transpose_15 = None + x_190 = torch._C._nn.linear( + x_189, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_189 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + mul_28 = x_191 * l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + x_191 = l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = None + x_192 = x_186 + mul_28 + x_186 = mul_28 = None + x_193 = torch.nn.functional.layer_norm( + x_192, + (1280,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_193 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_195 = torch._C._nn.gelu(x_194, approximate="none") + x_194 = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + mul_29 = x_198 * l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + x_198 = l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = None + x_199 = x_192 + mul_29 + x_192 = mul_29 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1280,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 257, 3, 16, 80) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_16.reshape(1, 257, 1280) + transpose_16 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + mul_30 = x_204 * l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + x_204 = l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = None + x_205 = x_199 + mul_30 + x_199 = mul_30 = None + x_206 = torch.nn.functional.layer_norm( + x_205, + (1280,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_207 = torch._C._nn.linear( + x_206, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_206 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_208 = torch._C._nn.gelu(x_207, approximate="none") + x_207 = None + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_209 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + mul_31 = x_211 * l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + x_211 = l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = None + x_212 = x_205 + mul_31 + x_205 = mul_31 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (1280,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_213 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 257, 3, 16, 80) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_214 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_214.transpose(1, 2) + x_214 = None + x_215 = transpose_17.reshape(1, 257, 1280) + transpose_17 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_215 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_217 = torch.nn.functional.dropout(x_216, 0.0, False, False) + x_216 = None + mul_32 = x_217 * l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + x_217 = l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = None + x_218 = x_212 + mul_32 + x_212 = mul_32 = None + x_219 = torch.nn.functional.layer_norm( + x_218, + (1280,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_219 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_221 = torch._C._nn.gelu(x_220, approximate="none") + x_220 = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + mul_33 = x_224 * l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + x_224 = l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = None + x_225 = x_218 + mul_33 + x_218 = mul_33 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (1280,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 257, 3, 16, 80) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_227 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_227.transpose(1, 2) + x_227 = None + x_228 = transpose_18.reshape(1, 257, 1280) + transpose_18 = None + x_229 = torch._C._nn.linear( + x_228, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_228 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + mul_34 = x_230 * l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + x_230 = l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = None + x_231 = x_225 + mul_34 + x_225 = mul_34 = None + x_232 = torch.nn.functional.layer_norm( + x_231, + (1280,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_232 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + mul_35 = x_237 * l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + x_237 = l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = None + x_238 = x_231 + mul_35 + x_231 = mul_35 = None + x_239 = torch.nn.functional.layer_norm( + x_238, + (1280,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_239, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_239 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 257, 3, 16, 80) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_19.reshape(1, 257, 1280) + transpose_19 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + mul_36 = x_243 * l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + x_243 = l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = None + x_244 = x_238 + mul_36 + x_238 = mul_36 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1280,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_247 = torch._C._nn.gelu(x_246, approximate="none") + x_246 = None + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_248 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + mul_37 = x_250 * l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + x_250 = l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = None + x_251 = x_244 + mul_37 + x_244 = mul_37 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1280,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 257, 3, 16, 80) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_253 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_253.transpose(1, 2) + x_253 = None + x_254 = transpose_20.reshape(1, 257, 1280) + transpose_20 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + mul_38 = x_256 * l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + x_256 = l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = None + x_257 = x_251 + mul_38 + x_251 = mul_38 = None + x_258 = torch.nn.functional.layer_norm( + x_257, + (1280,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_259 = torch._C._nn.linear( + x_258, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_258 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_260 = torch._C._nn.gelu(x_259, approximate="none") + x_259 = None + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_261 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + mul_39 = x_263 * l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + x_263 = l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = None + x_264 = x_257 + mul_39 + x_257 = mul_39 = None + x_265 = torch.nn.functional.layer_norm( + x_264, + (1280,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_265, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_265 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 257, 3, 16, 80) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_266 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_266.transpose(1, 2) + x_266 = None + x_267 = transpose_21.reshape(1, 257, 1280) + transpose_21 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_267 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + mul_40 = x_269 * l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + x_269 = l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = None + x_270 = x_264 + mul_40 + x_264 = mul_40 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (1280,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_273 = torch._C._nn.gelu(x_272, approximate="none") + x_272 = None + x_274 = torch.nn.functional.dropout(x_273, 0.0, False, False) + x_273 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_274 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_276 = torch.nn.functional.dropout(x_275, 0.0, False, False) + x_275 = None + mul_41 = x_276 * l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + x_276 = l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = None + x_277 = x_270 + mul_41 + x_270 = mul_41 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1280,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_278 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 257, 3, 16, 80) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_279 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_22.reshape(1, 257, 1280) + transpose_22 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_280 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + mul_42 = x_282 * l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + x_282 = l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = None + x_283 = x_277 + mul_42 + x_277 = mul_42 = None + x_284 = torch.nn.functional.layer_norm( + x_283, + (1280,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_285 = torch._C._nn.linear( + x_284, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_284 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_286 = torch._C._nn.gelu(x_285, approximate="none") + x_285 = None + x_287 = torch.nn.functional.dropout(x_286, 0.0, False, False) + x_286 = None + x_288 = torch._C._nn.linear( + x_287, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_287 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + mul_43 = x_289 * l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + x_289 = l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = None + x_290 = x_283 + mul_43 + x_283 = mul_43 = None + x_291 = torch.nn.functional.layer_norm( + x_290, + (1280,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_291, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_291 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 257, 3, 16, 80) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_292 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_292.transpose(1, 2) + x_292 = None + x_293 = transpose_23.reshape(1, 257, 1280) + transpose_23 = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + mul_44 = x_295 * l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + x_295 = l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = None + x_296 = x_290 + mul_44 + x_290 = mul_44 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1280,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + mul_45 = x_302 * l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + x_302 = l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = None + x_303 = x_296 + mul_45 + x_296 = mul_45 = None + x_304 = torch.nn.functional.layer_norm( + x_303, + (1280,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_304, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_304 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 257, 3, 16, 80) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_305 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_305.transpose(1, 2) + x_305 = None + x_306 = transpose_24.reshape(1, 257, 1280) + transpose_24 = None + x_307 = torch._C._nn.linear( + x_306, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_306 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + mul_46 = x_308 * l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + x_308 = l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = None + x_309 = x_303 + mul_46 + x_303 = mul_46 = None + x_310 = torch.nn.functional.layer_norm( + x_309, + (1280,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_311 = torch._C._nn.linear( + x_310, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_310 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_312 = torch._C._nn.gelu(x_311, approximate="none") + x_311 = None + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_313 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + mul_47 = x_315 * l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + x_315 = l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = None + x_316 = x_309 + mul_47 + x_309 = mul_47 = None + x_317 = torch.nn.functional.layer_norm( + x_316, + (1280,), + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + x_317, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_, + ) + x_317 = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_48 = linear_96.reshape(1, 257, 3, 16, 80) + linear_96 = None + qkv_24 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_24 = qkv_24.unbind(0) + qkv_24 = None + q_24 = unbind_24[0] + k_24 = unbind_24[1] + v_24 = unbind_24[2] + unbind_24 = None + x_318 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_25 = x_318.transpose(1, 2) + x_318 = None + x_319 = transpose_25.reshape(1, 257, 1280) + transpose_25 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_, + ) + x_319 = l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) = None + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + mul_48 = x_321 * l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ + x_321 = l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ = None + x_322 = x_316 + mul_48 + x_316 = mul_48 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1280,), + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_323 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_325 = torch._C._nn.gelu(x_324, approximate="none") + x_324 = None + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = torch._C._nn.linear( + x_326, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_326 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_328 = torch.nn.functional.dropout(x_327, 0.0, False, False) + x_327 = None + mul_49 = x_328 * l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ + x_328 = l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ = None + x_329 = x_322 + mul_49 + x_322 = mul_49 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1280,), + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) = None + linear_100 = torch._C._nn.linear( + x_330, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_, + ) + x_330 = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_50 = linear_100.reshape(1, 257, 3, 16, 80) + linear_100 = None + qkv_25 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_25 = qkv_25.unbind(0) + qkv_25 = None + q_25 = unbind_25[0] + k_25 = unbind_25[1] + v_25 = unbind_25[2] + unbind_25 = None + x_331 = torch._C._nn.scaled_dot_product_attention( + q_25, k_25, v_25, attn_mask=None, dropout_p=0.0 + ) + q_25 = k_25 = v_25 = None + transpose_26 = x_331.transpose(1, 2) + x_331 = None + x_332 = transpose_26.reshape(1, 257, 1280) + transpose_26 = None + x_333 = torch._C._nn.linear( + x_332, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_, + ) + x_332 = l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) = None + x_334 = torch.nn.functional.dropout(x_333, 0.0, False, False) + x_333 = None + mul_50 = x_334 * l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ + x_334 = l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ = None + x_335 = x_329 + mul_50 + x_329 = mul_50 = None + x_336 = torch.nn.functional.layer_norm( + x_335, + (1280,), + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) = None + x_337 = torch._C._nn.linear( + x_336, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_336 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_338 = torch._C._nn.gelu(x_337, approximate="none") + x_337 = None + x_339 = torch.nn.functional.dropout(x_338, 0.0, False, False) + x_338 = None + x_340 = torch._C._nn.linear( + x_339, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_339 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + mul_51 = x_341 * l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ + x_341 = l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ = None + x_342 = x_335 + mul_51 + x_335 = mul_51 = None + x_343 = torch.nn.functional.layer_norm( + x_342, + (1280,), + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) = None + linear_104 = torch._C._nn.linear( + x_343, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_, + ) + x_343 = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_52 = linear_104.reshape(1, 257, 3, 16, 80) + linear_104 = None + qkv_26 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_26 = qkv_26.unbind(0) + qkv_26 = None + q_26 = unbind_26[0] + k_26 = unbind_26[1] + v_26 = unbind_26[2] + unbind_26 = None + x_344 = torch._C._nn.scaled_dot_product_attention( + q_26, k_26, v_26, attn_mask=None, dropout_p=0.0 + ) + q_26 = k_26 = v_26 = None + transpose_27 = x_344.transpose(1, 2) + x_344 = None + x_345 = transpose_27.reshape(1, 257, 1280) + transpose_27 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_, + ) + x_345 = l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) = None + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + mul_52 = x_347 * l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ + x_347 = l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ = None + x_348 = x_342 + mul_52 + x_342 = mul_52 = None + x_349 = torch.nn.functional.layer_norm( + x_348, + (1280,), + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) = None + x_350 = torch._C._nn.linear( + x_349, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_349 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_351 = torch._C._nn.gelu(x_350, approximate="none") + x_350 = None + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_352 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + mul_53 = x_354 * l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ + x_354 = l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ = None + x_355 = x_348 + mul_53 + x_348 = mul_53 = None + x_356 = torch.nn.functional.layer_norm( + x_355, + (1280,), + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) = None + linear_108 = torch._C._nn.linear( + x_356, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_, + ) + x_356 = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_54 = linear_108.reshape(1, 257, 3, 16, 80) + linear_108 = None + qkv_27 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_27 = qkv_27.unbind(0) + qkv_27 = None + q_27 = unbind_27[0] + k_27 = unbind_27[1] + v_27 = unbind_27[2] + unbind_27 = None + x_357 = torch._C._nn.scaled_dot_product_attention( + q_27, k_27, v_27, attn_mask=None, dropout_p=0.0 + ) + q_27 = k_27 = v_27 = None + transpose_28 = x_357.transpose(1, 2) + x_357 = None + x_358 = transpose_28.reshape(1, 257, 1280) + transpose_28 = None + x_359 = torch._C._nn.linear( + x_358, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_, + ) + x_358 = l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + mul_54 = x_360 * l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ + x_360 = l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ = None + x_361 = x_355 + mul_54 + x_355 = mul_54 = None + x_362 = torch.nn.functional.layer_norm( + x_361, + (1280,), + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) = None + x_363 = torch._C._nn.linear( + x_362, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_, + ) + x_362 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_364 = torch._C._nn.gelu(x_363, approximate="none") + x_363 = None + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = torch._C._nn.linear( + x_365, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_, + ) + x_365 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_367 = torch.nn.functional.dropout(x_366, 0.0, False, False) + x_366 = None + mul_55 = x_367 * l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ + x_367 = l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ = None + x_368 = x_361 + mul_55 + x_361 = mul_55 = None + x_369 = torch.nn.functional.layer_norm( + x_368, + (1280,), + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) = None + linear_112 = torch._C._nn.linear( + x_369, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_, + ) + x_369 = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_56 = linear_112.reshape(1, 257, 3, 16, 80) + linear_112 = None + qkv_28 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_28 = qkv_28.unbind(0) + qkv_28 = None + q_28 = unbind_28[0] + k_28 = unbind_28[1] + v_28 = unbind_28[2] + unbind_28 = None + x_370 = torch._C._nn.scaled_dot_product_attention( + q_28, k_28, v_28, attn_mask=None, dropout_p=0.0 + ) + q_28 = k_28 = v_28 = None + transpose_29 = x_370.transpose(1, 2) + x_370 = None + x_371 = transpose_29.reshape(1, 257, 1280) + transpose_29 = None + x_372 = torch._C._nn.linear( + x_371, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_, + ) + x_371 = l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) = None + x_373 = torch.nn.functional.dropout(x_372, 0.0, False, False) + x_372 = None + mul_56 = x_373 * l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ + x_373 = l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ = None + x_374 = x_368 + mul_56 + x_368 = mul_56 = None + x_375 = torch.nn.functional.layer_norm( + x_374, + (1280,), + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) = None + x_376 = torch._C._nn.linear( + x_375, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_, + ) + x_375 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_377 = torch._C._nn.gelu(x_376, approximate="none") + x_376 = None + x_378 = torch.nn.functional.dropout(x_377, 0.0, False, False) + x_377 = None + x_379 = torch._C._nn.linear( + x_378, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_, + ) + x_378 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + mul_57 = x_380 * l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ + x_380 = l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ = None + x_381 = x_374 + mul_57 + x_374 = mul_57 = None + x_382 = torch.nn.functional.layer_norm( + x_381, + (1280,), + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) = None + linear_116 = torch._C._nn.linear( + x_382, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_, + ) + x_382 = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_58 = linear_116.reshape(1, 257, 3, 16, 80) + linear_116 = None + qkv_29 = reshape_58.permute(2, 0, 3, 1, 4) + reshape_58 = None + unbind_29 = qkv_29.unbind(0) + qkv_29 = None + q_29 = unbind_29[0] + k_29 = unbind_29[1] + v_29 = unbind_29[2] + unbind_29 = None + x_383 = torch._C._nn.scaled_dot_product_attention( + q_29, k_29, v_29, attn_mask=None, dropout_p=0.0 + ) + q_29 = k_29 = v_29 = None + transpose_30 = x_383.transpose(1, 2) + x_383 = None + x_384 = transpose_30.reshape(1, 257, 1280) + transpose_30 = None + x_385 = torch._C._nn.linear( + x_384, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_, + ) + x_384 = l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) = None + x_386 = torch.nn.functional.dropout(x_385, 0.0, False, False) + x_385 = None + mul_58 = x_386 * l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ + x_386 = l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ = None + x_387 = x_381 + mul_58 + x_381 = mul_58 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1280,), + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) = None + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_390 = torch._C._nn.gelu(x_389, approximate="none") + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + mul_59 = x_393 * l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ + x_393 = l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ = None + x_394 = x_387 + mul_59 + x_387 = mul_59 = None + x_395 = torch.nn.functional.layer_norm( + x_394, + (1280,), + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) = None + linear_120 = torch._C._nn.linear( + x_395, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_, + ) + x_395 = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_60 = linear_120.reshape(1, 257, 3, 16, 80) + linear_120 = None + qkv_30 = reshape_60.permute(2, 0, 3, 1, 4) + reshape_60 = None + unbind_30 = qkv_30.unbind(0) + qkv_30 = None + q_30 = unbind_30[0] + k_30 = unbind_30[1] + v_30 = unbind_30[2] + unbind_30 = None + x_396 = torch._C._nn.scaled_dot_product_attention( + q_30, k_30, v_30, attn_mask=None, dropout_p=0.0 + ) + q_30 = k_30 = v_30 = None + transpose_31 = x_396.transpose(1, 2) + x_396 = None + x_397 = transpose_31.reshape(1, 257, 1280) + transpose_31 = None + x_398 = torch._C._nn.linear( + x_397, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_, + ) + x_397 = l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) = None + x_399 = torch.nn.functional.dropout(x_398, 0.0, False, False) + x_398 = None + mul_60 = x_399 * l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ + x_399 = l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ = None + x_400 = x_394 + mul_60 + x_394 = mul_60 = None + x_401 = torch.nn.functional.layer_norm( + x_400, + (1280,), + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_, + ) + x_401 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_403 = torch._C._nn.gelu(x_402, approximate="none") + x_402 = None + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = torch._C._nn.linear( + x_404, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_, + ) + x_404 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_406 = torch.nn.functional.dropout(x_405, 0.0, False, False) + x_405 = None + mul_61 = x_406 * l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ + x_406 = l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ = None + x_407 = x_400 + mul_61 + x_400 = mul_61 = None + x_408 = torch.nn.functional.layer_norm( + x_407, + (1280,), + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) = None + linear_124 = torch._C._nn.linear( + x_408, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_, + ) + x_408 = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_62 = linear_124.reshape(1, 257, 3, 16, 80) + linear_124 = None + qkv_31 = reshape_62.permute(2, 0, 3, 1, 4) + reshape_62 = None + unbind_31 = qkv_31.unbind(0) + qkv_31 = None + q_31 = unbind_31[0] + k_31 = unbind_31[1] + v_31 = unbind_31[2] + unbind_31 = None + x_409 = torch._C._nn.scaled_dot_product_attention( + q_31, k_31, v_31, attn_mask=None, dropout_p=0.0 + ) + q_31 = k_31 = v_31 = None + transpose_32 = x_409.transpose(1, 2) + x_409 = None + x_410 = transpose_32.reshape(1, 257, 1280) + transpose_32 = None + x_411 = torch._C._nn.linear( + x_410, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_, + ) + x_410 = l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) = None + x_412 = torch.nn.functional.dropout(x_411, 0.0, False, False) + x_411 = None + mul_62 = x_412 * l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ + x_412 = l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ = None + x_413 = x_407 + mul_62 + x_407 = mul_62 = None + x_414 = torch.nn.functional.layer_norm( + x_413, + (1280,), + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) = None + x_415 = torch._C._nn.linear( + x_414, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_, + ) + x_414 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_416 = torch._C._nn.gelu(x_415, approximate="none") + x_415 = None + x_417 = torch.nn.functional.dropout(x_416, 0.0, False, False) + x_416 = None + x_418 = torch._C._nn.linear( + x_417, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_, + ) + x_417 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_419 = torch.nn.functional.dropout(x_418, 0.0, False, False) + x_418 = None + mul_63 = x_419 * l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ + x_419 = l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ = None + x_420 = x_413 + mul_63 + x_413 = mul_63 = None + x_421 = torch.nn.functional.layer_norm( + x_420, + (1280,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_420 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_422 = x_421[(slice(None, None, None), 0)] + x_421 = None + x_423 = torch.nn.functional.dropout(x_422, 0.0, False, False) + x_422 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_423 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_424,) diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..35cfbe679 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py @@ -0,0 +1,4676 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1280, 3, 14, 14] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 256, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..a9553b407 --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +7adc40d79aba5a23828e77fe6b175c6b16b27f015e71f10b3c9e91352272f0ec \ No newline at end of file diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_large_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_large_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/model.py b/samples/timm/deit3_large_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..8be144f1b --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/model.py @@ -0,0 +1,3620 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_161 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + mul_24 = x_165 * l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + x_165 = l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = None + x_166 = x_160 + mul_24 + x_160 = mul_24 = None + x_167 = torch.nn.functional.layer_norm( + x_166, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + x_167, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_167 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_25 = x_172 * l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + x_172 = l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = None + x_173 = x_166 + mul_25 + x_166 = mul_25 = None + x_174 = torch.nn.functional.layer_norm( + x_173, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_174 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_175 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_175.transpose(1, 2) + x_175 = None + x_176 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_176 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_178 = torch.nn.functional.dropout(x_177, 0.0, False, False) + x_177 = None + mul_26 = x_178 * l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + x_178 = l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = None + x_179 = x_173 + mul_26 + x_173 = mul_26 = None + x_180 = torch.nn.functional.layer_norm( + x_179, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_182 = torch._C._nn.gelu(x_181, approximate="none") + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + mul_27 = x_185 * l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + x_185 = l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = None + x_186 = x_179 + mul_27 + x_179 = mul_27 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_187, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_187 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_188 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_188.transpose(1, 2) + x_188 = None + x_189 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_190 = torch._C._nn.linear( + x_189, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_189 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + mul_28 = x_191 * l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + x_191 = l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = None + x_192 = x_186 + mul_28 + x_186 = mul_28 = None + x_193 = torch.nn.functional.layer_norm( + x_192, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_193 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_195 = torch._C._nn.gelu(x_194, approximate="none") + x_194 = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + mul_29 = x_198 * l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + x_198 = l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = None + x_199 = x_192 + mul_29 + x_192 = mul_29 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + mul_30 = x_204 * l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + x_204 = l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = None + x_205 = x_199 + mul_30 + x_199 = mul_30 = None + x_206 = torch.nn.functional.layer_norm( + x_205, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_207 = torch._C._nn.linear( + x_206, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_206 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_208 = torch._C._nn.gelu(x_207, approximate="none") + x_207 = None + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_209 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + mul_31 = x_211 * l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + x_211 = l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = None + x_212 = x_205 + mul_31 + x_205 = mul_31 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_213 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_214 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_214.transpose(1, 2) + x_214 = None + x_215 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_215 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_217 = torch.nn.functional.dropout(x_216, 0.0, False, False) + x_216 = None + mul_32 = x_217 * l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + x_217 = l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = None + x_218 = x_212 + mul_32 + x_212 = mul_32 = None + x_219 = torch.nn.functional.layer_norm( + x_218, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_219 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_221 = torch._C._nn.gelu(x_220, approximate="none") + x_220 = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + mul_33 = x_224 * l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + x_224 = l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = None + x_225 = x_218 + mul_33 + x_218 = mul_33 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_227 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_227.transpose(1, 2) + x_227 = None + x_228 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_229 = torch._C._nn.linear( + x_228, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_228 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + mul_34 = x_230 * l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + x_230 = l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = None + x_231 = x_225 + mul_34 + x_225 = mul_34 = None + x_232 = torch.nn.functional.layer_norm( + x_231, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_232 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + mul_35 = x_237 * l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + x_237 = l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = None + x_238 = x_231 + mul_35 + x_231 = mul_35 = None + x_239 = torch.nn.functional.layer_norm( + x_238, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_239, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_239 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + mul_36 = x_243 * l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + x_243 = l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = None + x_244 = x_238 + mul_36 + x_238 = mul_36 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_247 = torch._C._nn.gelu(x_246, approximate="none") + x_246 = None + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_248 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + mul_37 = x_250 * l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + x_250 = l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = None + x_251 = x_244 + mul_37 + x_244 = mul_37 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_253 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_253.transpose(1, 2) + x_253 = None + x_254 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + mul_38 = x_256 * l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + x_256 = l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = None + x_257 = x_251 + mul_38 + x_251 = mul_38 = None + x_258 = torch.nn.functional.layer_norm( + x_257, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_259 = torch._C._nn.linear( + x_258, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_258 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_260 = torch._C._nn.gelu(x_259, approximate="none") + x_259 = None + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_261 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + mul_39 = x_263 * l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + x_263 = l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = None + x_264 = x_257 + mul_39 + x_257 = mul_39 = None + x_265 = torch.nn.functional.layer_norm( + x_264, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_265, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_265 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_266 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_266.transpose(1, 2) + x_266 = None + x_267 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_267 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + mul_40 = x_269 * l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + x_269 = l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = None + x_270 = x_264 + mul_40 + x_264 = mul_40 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_273 = torch._C._nn.gelu(x_272, approximate="none") + x_272 = None + x_274 = torch.nn.functional.dropout(x_273, 0.0, False, False) + x_273 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_274 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_276 = torch.nn.functional.dropout(x_275, 0.0, False, False) + x_275 = None + mul_41 = x_276 * l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + x_276 = l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = None + x_277 = x_270 + mul_41 + x_270 = mul_41 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_278 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_279 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_280 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + mul_42 = x_282 * l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + x_282 = l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = None + x_283 = x_277 + mul_42 + x_277 = mul_42 = None + x_284 = torch.nn.functional.layer_norm( + x_283, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_285 = torch._C._nn.linear( + x_284, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_284 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_286 = torch._C._nn.gelu(x_285, approximate="none") + x_285 = None + x_287 = torch.nn.functional.dropout(x_286, 0.0, False, False) + x_286 = None + x_288 = torch._C._nn.linear( + x_287, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_287 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + mul_43 = x_289 * l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + x_289 = l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = None + x_290 = x_283 + mul_43 + x_283 = mul_43 = None + x_291 = torch.nn.functional.layer_norm( + x_290, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_291, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_291 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_292 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_292.transpose(1, 2) + x_292 = None + x_293 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + mul_44 = x_295 * l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + x_295 = l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = None + x_296 = x_290 + mul_44 + x_290 = mul_44 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + mul_45 = x_302 * l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + x_302 = l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = None + x_303 = x_296 + mul_45 + x_296 = mul_45 = None + x_304 = torch.nn.functional.layer_norm( + x_303, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_304, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_304 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_305 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_305.transpose(1, 2) + x_305 = None + x_306 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_307 = torch._C._nn.linear( + x_306, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_306 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + mul_46 = x_308 * l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + x_308 = l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = None + x_309 = x_303 + mul_46 + x_303 = mul_46 = None + x_310 = torch.nn.functional.layer_norm( + x_309, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_311 = torch._C._nn.linear( + x_310, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_310 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_312 = torch._C._nn.gelu(x_311, approximate="none") + x_311 = None + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_313 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + mul_47 = x_315 * l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + x_315 = l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = None + x_316 = x_309 + mul_47 + x_309 = mul_47 = None + x_317 = torch.nn.functional.layer_norm( + x_316, + (1024,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_316 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_318 = x_317[(slice(None, None, None), 0)] + x_317 = None + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_319 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_320,) diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..5582416ce --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,3524 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..916e637e1 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +c3ddd27825c23ebd4fdf6c5ca7a97ccd364dcb24ed54194e97395690eedf7055 \ No newline at end of file diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..0d76c5659 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (512,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 8, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 512) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (512,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (512,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 8, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 512) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (512,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (512,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 8, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 512) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (512,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (512,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 8, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 512) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (512,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (512,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 8, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 512) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (512,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (512,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 8, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 512) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (512,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (512,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 8, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 512) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (512,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (512,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 8, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 512) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (512,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (512,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 8, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 512) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (512,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (512,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 8, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 512) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (512,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 8, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 512) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (512,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (512,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 8, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 512) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (512,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..d43437c23 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [512, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..36014e957 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +528fbf8dda6333381aba8de97f9d882159ac642553a4ac9d788244a60b2425d9 \ No newline at end of file diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_small_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_small_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/model.py b/samples/timm/deit3_small_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..ea4b9f7d2 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..7cab472b3 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..a70a97d2d --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +c83bcf76f42e39c7be55f30e59ef7cef3a5854c9abd5fd91ef151a609a558094 \ No newline at end of file diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..878dfe9b3 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..1ff0dd1d2 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..40fb26dcd --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +1d0ef300e50462f683e3b88ea3674a392cb6ecc17bc7dd39f8e2c65367e829a5 \ No newline at end of file diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_base_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_base_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/model.py b/samples/timm/deit_base_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..49c6c7e90 --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..e6b299a89 --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..f602a0660 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +a6fd91eca0badc635ce20aead25e47f6b7aaeef35de34742b760572723b41494 \ No newline at end of file diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..9d7fb709b --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..251165b15 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..4d06412e6 --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +dd39f4fd656faad35082e94c9b6536c58e6fa1fde59fbf16b0e8d53977453a6f \ No newline at end of file diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_small_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_small_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/model.py b/samples/timm/deit_small_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..d40e5970e --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..f6c0831e3 --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..9c0d5dce4 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +8f0e9cfd2900cb44268638505c1031a525ea15e20dbf26a3feddfa71f0f303b5 \ No newline at end of file diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..df876a6e6 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 3, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 192) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 192) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 192) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 192) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 192) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 3, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 192) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 3, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 192) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 3, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 192) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 3, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 192) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 3, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 192) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 3, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 192) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 3, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 192) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..93d831932 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 000000000..b269f0719 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +280ac56eaaf4721e2023e2a905b9645fb0eea6be3fa203adaded06373c53b6b5 \ No newline at end of file diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py new file mode 100644 index 000000000..9b99f4fff --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 3, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 192) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 192) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 192) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 192) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 192) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 3, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 192) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 3, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 192) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 3, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 192) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 3, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 192) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 3, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 192) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 3, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 192) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 3, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 192) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 000000000..a76a4e305 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet121.ra_in1k/graph_hash.txt b/samples/timm/densenet121.ra_in1k/graph_hash.txt new file mode 100644 index 000000000..4aefe5c6c --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +7d877cb38ecbe22220db7dc1f7b2d0cf9438bfb4f70998ecee1ae10941e4eeb2 \ No newline at end of file diff --git a/samples/timm/densenet121.ra_in1k/graph_net.json b/samples/timm/densenet121.ra_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet121.ra_in1k/input_meta.py b/samples/timm/densenet121.ra_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet121.ra_in1k/input_tensor_constraints.py b/samples/timm/densenet121.ra_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet121.ra_in1k/model.py b/samples/timm/densenet121.ra_in1k/model.py new file mode 100644 index 000000000..7aaea4dc3 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/model.py @@ -0,0 +1,4977 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = new_features_38 = new_features_39 = new_features_40 = new_features_41 = None + x_174 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + input_10 = torch.conv2d( + x_175, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_42 = torch.cat([input_11], 1) + x_176 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + bottleneck_output_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + new_features_42 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat([input_11, new_features_42], 1) + x_180 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + bottleneck_output_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_182 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + new_features_43 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [input_11, new_features_42, new_features_43], 1 + ) + x_184 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + bottleneck_output_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_186 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + new_features_44 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [input_11, new_features_42, new_features_43, new_features_44], 1 + ) + x_188 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + bottleneck_output_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + new_features_45 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_192 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + bottleneck_output_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_194 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + new_features_46 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_196 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + bottleneck_output_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_198 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + new_features_47 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_200 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + bottleneck_output_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + new_features_48 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_204 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + bottleneck_output_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_206 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + new_features_49 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_208 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + bottleneck_output_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_210 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + new_features_50 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_212 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + bottleneck_output_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_214 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + new_features_51 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_216 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + bottleneck_output_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_218 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + new_features_52 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_220 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + bottleneck_output_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + new_features_53 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + input_11 = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = new_features_54 = new_features_55 = new_features_56 = new_features_57 = None + x_240 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + x_242 = torch.nn.functional.adaptive_avg_pool2d(x_241, 1) + x_241 = None + x_243 = x_242.flatten(1, -1) + x_242 = None + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = torch._C._nn.linear( + x_244, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_244 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_245,) diff --git a/samples/timm/densenet121.ra_in1k/weight_meta.py b/samples/timm/densenet121.ra_in1k/weight_meta.py new file mode 100644 index 000000000..7d3a290ff --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/weight_meta.py @@ -0,0 +1,6090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.060 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet161.tv_in1k/graph_hash.txt b/samples/timm/densenet161.tv_in1k/graph_hash.txt new file mode 100644 index 000000000..5b716bb92 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +4cef2f10705d139d718cf68cff2ac72ea0f705b85b19a9b403273b7308adb826 \ No newline at end of file diff --git a/samples/timm/densenet161.tv_in1k/graph_net.json b/samples/timm/densenet161.tv_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet161.tv_in1k/input_meta.py b/samples/timm/densenet161.tv_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet161.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet161.tv_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet161.tv_in1k/model.py b/samples/timm/densenet161.tv_in1k/model.py new file mode 100644 index 000000000..b0f34c0e7 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/model.py @@ -0,0 +1,6987 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_206 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + bottleneck_output_50 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + new_features_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_210 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + bottleneck_output_51 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + new_features_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_214 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + bottleneck_output_52 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = (None) + x_216 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + new_features_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_218 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + bottleneck_output_53 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + new_features_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = new_features_50 = new_features_51 = new_features_52 = new_features_53 = None + x_222 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + input_10 = torch.conv2d( + x_223, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_223 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_54 = torch.cat([input_11], 1) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat([input_11, new_features_54], 1) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [input_11, new_features_54, new_features_55], 1 + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [input_11, new_features_54, new_features_55, new_features_56], 1 + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_240 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + bottleneck_output_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_242 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + new_features_58 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_244 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + bottleneck_output_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_246 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + new_features_59 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_248 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + bottleneck_output_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_250 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + new_features_60 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_252 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + bottleneck_output_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_254 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + new_features_61 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_256 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + bottleneck_output_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_258 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + new_features_62 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_260 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + bottleneck_output_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_262 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + new_features_63 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_264 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + bottleneck_output_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_266 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + new_features_64 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_268 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + bottleneck_output_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_270 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + new_features_65 = torch.conv2d( + x_271, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_271 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_66 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + ], + 1, + ) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + ], + 1, + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + ], + 1, + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + input_11 = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = ( + new_features_62 + ) = ( + new_features_63 + ) = ( + new_features_64 + ) = ( + new_features_65 + ) = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = new_features_74 = new_features_75 = new_features_76 = new_features_77 = None + x_320 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + x_322 = torch.nn.functional.adaptive_avg_pool2d(x_321, 1) + x_321 = None + x_323 = x_322.flatten(1, -1) + x_322 = None + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_324 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_325,) diff --git a/samples/timm/densenet161.tv_in1k/weight_meta.py b/samples/timm/densenet161.tv_in1k/weight_meta.py new file mode 100644 index 000000000..7efccecee --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/weight_meta.py @@ -0,0 +1,8090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 144, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.118 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 336, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 336, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 432, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 528, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 624, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 720, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [384, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 432, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 528, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 624, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 720, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 816, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 912, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [192, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [192, 1008, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [192, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [192, 1104, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [192, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [192, 1200, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [192, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [192, 1296, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [192, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [192, 1392, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [192, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [192, 1488, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [192, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [192, 1584, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [192, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [192, 1680, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [192, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [192, 1776, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [192, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [192, 1872, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_" + shape = [192, 1920, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_" + shape = [192, 1968, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_" + shape = [192, 2016, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_" + shape = [192, 2064, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [1056, 2112, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 1104, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 1200, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 1296, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 1392, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 1488, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 1584, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [192, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [192, 1680, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [192, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [192, 1776, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [192, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [192, 1872, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [192, 1920, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [192, 1968, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [192, 2016, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [192, 2064, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [192, 2112, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [192, 2160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.012 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet169.tv_in1k/graph_hash.txt b/samples/timm/densenet169.tv_in1k/graph_hash.txt new file mode 100644 index 000000000..9ccec065d --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +20ec56b7a5ce2dcf23a0114ba99191eaa451499b4406b1eec91ea0b1bd44dfac \ No newline at end of file diff --git a/samples/timm/densenet169.tv_in1k/graph_net.json b/samples/timm/densenet169.tv_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet169.tv_in1k/input_meta.py b/samples/timm/densenet169.tv_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet169.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet169.tv_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet169.tv_in1k/model.py b/samples/timm/densenet169.tv_in1k/model.py new file mode 100644 index 000000000..ab882eb05 --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/model.py @@ -0,0 +1,7373 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = new_features_46 = new_features_47 = new_features_48 = new_features_49 = None + x_206 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + input_10 = torch.conv2d( + x_207, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_50 = torch.cat([input_11], 1) + x_208 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + bottleneck_output_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_210 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + new_features_50 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat([input_11, new_features_50], 1) + x_212 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + bottleneck_output_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_214 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + new_features_51 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [input_11, new_features_50, new_features_51], 1 + ) + x_216 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + bottleneck_output_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_218 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + new_features_52 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [input_11, new_features_50, new_features_51, new_features_52], 1 + ) + x_220 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + bottleneck_output_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + new_features_53 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_240 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + bottleneck_output_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_242 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + new_features_58 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_244 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + bottleneck_output_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_246 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + new_features_59 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_248 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + bottleneck_output_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_250 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + new_features_60 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_252 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + bottleneck_output_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_254 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + new_features_61 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_256 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + bottleneck_output_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_258 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + new_features_62 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_260 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + bottleneck_output_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_262 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + new_features_63 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_264 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + bottleneck_output_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_266 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + new_features_64 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_268 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + bottleneck_output_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_270 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + new_features_65 = torch.conv2d( + x_271, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_271 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_66 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + ], + 1, + ) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + ], + 1, + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + ], + 1, + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_78 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + x_320 = torch.nn.functional.batch_norm( + concated_features_78, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_78 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + bottleneck_output_78 = torch.conv2d( + x_321, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_321 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_322 = torch.nn.functional.batch_norm( + bottleneck_output_78, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_78 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_323 = torch.nn.functional.relu(x_322, inplace=True) + x_322 = None + new_features_78 = torch.conv2d( + x_323, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_323 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_79 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + ], + 1, + ) + x_324 = torch.nn.functional.batch_norm( + concated_features_79, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_79 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_325 = torch.nn.functional.relu(x_324, inplace=True) + x_324 = None + bottleneck_output_79 = torch.conv2d( + x_325, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_325 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_326 = torch.nn.functional.batch_norm( + bottleneck_output_79, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_79 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_327 = torch.nn.functional.relu(x_326, inplace=True) + x_326 = None + new_features_79 = torch.conv2d( + x_327, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_327 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_80 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + ], + 1, + ) + x_328 = torch.nn.functional.batch_norm( + concated_features_80, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_80 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_329 = torch.nn.functional.relu(x_328, inplace=True) + x_328 = None + bottleneck_output_80 = torch.conv2d( + x_329, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_329 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_330 = torch.nn.functional.batch_norm( + bottleneck_output_80, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_80 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_331 = torch.nn.functional.relu(x_330, inplace=True) + x_330 = None + new_features_80 = torch.conv2d( + x_331, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_331 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_81 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + ], + 1, + ) + x_332 = torch.nn.functional.batch_norm( + concated_features_81, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_81 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_333 = torch.nn.functional.relu(x_332, inplace=True) + x_332 = None + bottleneck_output_81 = torch.conv2d( + x_333, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_333 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_334 = torch.nn.functional.batch_norm( + bottleneck_output_81, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_81 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_335 = torch.nn.functional.relu(x_334, inplace=True) + x_334 = None + new_features_81 = torch.conv2d( + x_335, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_335 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + ], + 1, + ) + input_11 = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = ( + new_features_62 + ) = ( + new_features_63 + ) = ( + new_features_64 + ) = ( + new_features_65 + ) = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = ( + new_features_74 + ) = ( + new_features_75 + ) = ( + new_features_76 + ) = ( + new_features_77 + ) = new_features_78 = new_features_79 = new_features_80 = new_features_81 = None + x_336 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_337 = torch.nn.functional.relu(x_336, inplace=True) + x_336 = None + x_338 = torch.nn.functional.adaptive_avg_pool2d(x_337, 1) + x_337 = None + x_339 = x_338.flatten(1, -1) + x_338 = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = torch._C._nn.linear( + x_340, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_340 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_341,) diff --git a/samples/timm/densenet169.tv_in1k/weight_meta.py b/samples/timm/densenet169.tv_in1k/weight_meta.py new file mode 100644 index 000000000..4c5dd6bab --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/weight_meta.py @@ -0,0 +1,8490 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.119 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.073 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [640, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1664] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet201.tv_in1k/graph_hash.txt b/samples/timm/densenet201.tv_in1k/graph_hash.txt new file mode 100644 index 000000000..d8dae0872 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +36ec146b47b470780e08cd74b50727801ccadde5421d7ed89ea47ca7ff326e59 \ No newline at end of file diff --git a/samples/timm/densenet201.tv_in1k/graph_net.json b/samples/timm/densenet201.tv_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet201.tv_in1k/input_meta.py b/samples/timm/densenet201.tv_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet201.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet201.tv_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenet201.tv_in1k/model.py b/samples/timm/densenet201.tv_in1k/model.py new file mode 100644 index 000000000..3861018bf --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/model.py @@ -0,0 +1,9205 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_206 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + bottleneck_output_50 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + new_features_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_210 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + bottleneck_output_51 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + new_features_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_214 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + bottleneck_output_52 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = (None) + x_216 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + new_features_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_218 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + bottleneck_output_53 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + new_features_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_222 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + bottleneck_output_54 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ = (None) + x_224 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + new_features_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_226 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + bottleneck_output_55 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + new_features_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_230 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + bottleneck_output_56 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + new_features_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_234 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + bottleneck_output_57 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ = (None) + x_236 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + new_features_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_238 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + bottleneck_output_58 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ = (None) + x_240 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + new_features_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_242 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + bottleneck_output_59 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ = (None) + x_244 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + new_features_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_246 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + bottleneck_output_60 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ = (None) + x_248 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + new_features_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_250 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + bottleneck_output_61 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ = (None) + x_252 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + new_features_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_254 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + bottleneck_output_62 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ = (None) + x_256 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + new_features_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_258 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + bottleneck_output_63 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ = (None) + x_260 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + new_features_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_262 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + bottleneck_output_64 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ = (None) + x_264 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + new_features_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_266 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + bottleneck_output_65 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ = (None) + x_268 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + new_features_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = new_features_62 = new_features_63 = new_features_64 = new_features_65 = None + x_270 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + input_10 = torch.conv2d( + x_271, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_271 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_66 = torch.cat([input_11], 1) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat([input_11, new_features_66], 1) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [input_11, new_features_66, new_features_67], 1 + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [input_11, new_features_66, new_features_67, new_features_68], 1 + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_78 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + x_320 = torch.nn.functional.batch_norm( + concated_features_78, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_78 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + bottleneck_output_78 = torch.conv2d( + x_321, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_321 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_322 = torch.nn.functional.batch_norm( + bottleneck_output_78, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_78 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_323 = torch.nn.functional.relu(x_322, inplace=True) + x_322 = None + new_features_78 = torch.conv2d( + x_323, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_323 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_79 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + ], + 1, + ) + x_324 = torch.nn.functional.batch_norm( + concated_features_79, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_79 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_325 = torch.nn.functional.relu(x_324, inplace=True) + x_324 = None + bottleneck_output_79 = torch.conv2d( + x_325, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_325 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_326 = torch.nn.functional.batch_norm( + bottleneck_output_79, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_79 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_327 = torch.nn.functional.relu(x_326, inplace=True) + x_326 = None + new_features_79 = torch.conv2d( + x_327, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_327 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_80 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + ], + 1, + ) + x_328 = torch.nn.functional.batch_norm( + concated_features_80, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_80 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_329 = torch.nn.functional.relu(x_328, inplace=True) + x_328 = None + bottleneck_output_80 = torch.conv2d( + x_329, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_329 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_330 = torch.nn.functional.batch_norm( + bottleneck_output_80, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_80 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_331 = torch.nn.functional.relu(x_330, inplace=True) + x_330 = None + new_features_80 = torch.conv2d( + x_331, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_331 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_81 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + ], + 1, + ) + x_332 = torch.nn.functional.batch_norm( + concated_features_81, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_81 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_333 = torch.nn.functional.relu(x_332, inplace=True) + x_332 = None + bottleneck_output_81 = torch.conv2d( + x_333, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_333 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_334 = torch.nn.functional.batch_norm( + bottleneck_output_81, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_81 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_335 = torch.nn.functional.relu(x_334, inplace=True) + x_334 = None + new_features_81 = torch.conv2d( + x_335, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_335 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_82 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + ], + 1, + ) + x_336 = torch.nn.functional.batch_norm( + concated_features_82, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_82 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_337 = torch.nn.functional.relu(x_336, inplace=True) + x_336 = None + bottleneck_output_82 = torch.conv2d( + x_337, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_337 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_338 = torch.nn.functional.batch_norm( + bottleneck_output_82, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_82 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_339 = torch.nn.functional.relu(x_338, inplace=True) + x_338 = None + new_features_82 = torch.conv2d( + x_339, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_339 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_83 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + ], + 1, + ) + x_340 = torch.nn.functional.batch_norm( + concated_features_83, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_83 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_341 = torch.nn.functional.relu(x_340, inplace=True) + x_340 = None + bottleneck_output_83 = torch.conv2d( + x_341, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_341 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_342 = torch.nn.functional.batch_norm( + bottleneck_output_83, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_83 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_343 = torch.nn.functional.relu(x_342, inplace=True) + x_342 = None + new_features_83 = torch.conv2d( + x_343, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_343 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_84 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + ], + 1, + ) + x_344 = torch.nn.functional.batch_norm( + concated_features_84, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_84 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_345 = torch.nn.functional.relu(x_344, inplace=True) + x_344 = None + bottleneck_output_84 = torch.conv2d( + x_345, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_345 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_346 = torch.nn.functional.batch_norm( + bottleneck_output_84, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_84 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_347 = torch.nn.functional.relu(x_346, inplace=True) + x_346 = None + new_features_84 = torch.conv2d( + x_347, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_347 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_85 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + ], + 1, + ) + x_348 = torch.nn.functional.batch_norm( + concated_features_85, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_85 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_349 = torch.nn.functional.relu(x_348, inplace=True) + x_348 = None + bottleneck_output_85 = torch.conv2d( + x_349, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_349 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_350 = torch.nn.functional.batch_norm( + bottleneck_output_85, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_85 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_351 = torch.nn.functional.relu(x_350, inplace=True) + x_350 = None + new_features_85 = torch.conv2d( + x_351, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_351 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_86 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + ], + 1, + ) + x_352 = torch.nn.functional.batch_norm( + concated_features_86, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_86 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_353 = torch.nn.functional.relu(x_352, inplace=True) + x_352 = None + bottleneck_output_86 = torch.conv2d( + x_353, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_353 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_354 = torch.nn.functional.batch_norm( + bottleneck_output_86, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_86 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_355 = torch.nn.functional.relu(x_354, inplace=True) + x_354 = None + new_features_86 = torch.conv2d( + x_355, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_355 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_87 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + ], + 1, + ) + x_356 = torch.nn.functional.batch_norm( + concated_features_87, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_87 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_357 = torch.nn.functional.relu(x_356, inplace=True) + x_356 = None + bottleneck_output_87 = torch.conv2d( + x_357, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_357 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_358 = torch.nn.functional.batch_norm( + bottleneck_output_87, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_87 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_359 = torch.nn.functional.relu(x_358, inplace=True) + x_358 = None + new_features_87 = torch.conv2d( + x_359, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_359 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_88 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + ], + 1, + ) + x_360 = torch.nn.functional.batch_norm( + concated_features_88, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_88 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_361 = torch.nn.functional.relu(x_360, inplace=True) + x_360 = None + bottleneck_output_88 = torch.conv2d( + x_361, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_361 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_362 = torch.nn.functional.batch_norm( + bottleneck_output_88, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_88 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_363 = torch.nn.functional.relu(x_362, inplace=True) + x_362 = None + new_features_88 = torch.conv2d( + x_363, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_363 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_89 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + ], + 1, + ) + x_364 = torch.nn.functional.batch_norm( + concated_features_89, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_89 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_365 = torch.nn.functional.relu(x_364, inplace=True) + x_364 = None + bottleneck_output_89 = torch.conv2d( + x_365, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_365 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_366 = torch.nn.functional.batch_norm( + bottleneck_output_89, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_89 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_367 = torch.nn.functional.relu(x_366, inplace=True) + x_366 = None + new_features_89 = torch.conv2d( + x_367, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_367 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_90 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + ], + 1, + ) + x_368 = torch.nn.functional.batch_norm( + concated_features_90, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_90 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_369 = torch.nn.functional.relu(x_368, inplace=True) + x_368 = None + bottleneck_output_90 = torch.conv2d( + x_369, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_369 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_370 = torch.nn.functional.batch_norm( + bottleneck_output_90, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_90 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_371 = torch.nn.functional.relu(x_370, inplace=True) + x_370 = None + new_features_90 = torch.conv2d( + x_371, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_371 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_91 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + ], + 1, + ) + x_372 = torch.nn.functional.batch_norm( + concated_features_91, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_91 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_373 = torch.nn.functional.relu(x_372, inplace=True) + x_372 = None + bottleneck_output_91 = torch.conv2d( + x_373, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_373 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_374 = torch.nn.functional.batch_norm( + bottleneck_output_91, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_91 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_375 = torch.nn.functional.relu(x_374, inplace=True) + x_374 = None + new_features_91 = torch.conv2d( + x_375, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_375 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_92 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + ], + 1, + ) + x_376 = torch.nn.functional.batch_norm( + concated_features_92, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_92 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_377 = torch.nn.functional.relu(x_376, inplace=True) + x_376 = None + bottleneck_output_92 = torch.conv2d( + x_377, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_377 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_378 = torch.nn.functional.batch_norm( + bottleneck_output_92, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_92 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_379 = torch.nn.functional.relu(x_378, inplace=True) + x_378 = None + new_features_92 = torch.conv2d( + x_379, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_379 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_93 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + ], + 1, + ) + x_380 = torch.nn.functional.batch_norm( + concated_features_93, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_93 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_381 = torch.nn.functional.relu(x_380, inplace=True) + x_380 = None + bottleneck_output_93 = torch.conv2d( + x_381, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_381 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_382 = torch.nn.functional.batch_norm( + bottleneck_output_93, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_93 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_383 = torch.nn.functional.relu(x_382, inplace=True) + x_382 = None + new_features_93 = torch.conv2d( + x_383, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_383 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_94 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + ], + 1, + ) + x_384 = torch.nn.functional.batch_norm( + concated_features_94, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_94 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_385 = torch.nn.functional.relu(x_384, inplace=True) + x_384 = None + bottleneck_output_94 = torch.conv2d( + x_385, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_385 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_386 = torch.nn.functional.batch_norm( + bottleneck_output_94, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_94 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_387 = torch.nn.functional.relu(x_386, inplace=True) + x_386 = None + new_features_94 = torch.conv2d( + x_387, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_387 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_95 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + ], + 1, + ) + x_388 = torch.nn.functional.batch_norm( + concated_features_95, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_95 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_389 = torch.nn.functional.relu(x_388, inplace=True) + x_388 = None + bottleneck_output_95 = torch.conv2d( + x_389, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_389 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_390 = torch.nn.functional.batch_norm( + bottleneck_output_95, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_95 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_391 = torch.nn.functional.relu(x_390, inplace=True) + x_390 = None + new_features_95 = torch.conv2d( + x_391, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_391 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_96 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + ], + 1, + ) + x_392 = torch.nn.functional.batch_norm( + concated_features_96, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_96 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_393 = torch.nn.functional.relu(x_392, inplace=True) + x_392 = None + bottleneck_output_96 = torch.conv2d( + x_393, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_393 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_394 = torch.nn.functional.batch_norm( + bottleneck_output_96, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_96 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_395 = torch.nn.functional.relu(x_394, inplace=True) + x_394 = None + new_features_96 = torch.conv2d( + x_395, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_395 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_97 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + new_features_96, + ], + 1, + ) + x_396 = torch.nn.functional.batch_norm( + concated_features_97, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_97 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_397 = torch.nn.functional.relu(x_396, inplace=True) + x_396 = None + bottleneck_output_97 = torch.conv2d( + x_397, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_397 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_398 = torch.nn.functional.batch_norm( + bottleneck_output_97, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_97 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_399 = torch.nn.functional.relu(x_398, inplace=True) + x_398 = None + new_features_97 = torch.conv2d( + x_399, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_399 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + new_features_96, + new_features_97, + ], + 1, + ) + input_11 = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = ( + new_features_74 + ) = ( + new_features_75 + ) = ( + new_features_76 + ) = ( + new_features_77 + ) = ( + new_features_78 + ) = ( + new_features_79 + ) = ( + new_features_80 + ) = ( + new_features_81 + ) = ( + new_features_82 + ) = ( + new_features_83 + ) = ( + new_features_84 + ) = ( + new_features_85 + ) = ( + new_features_86 + ) = ( + new_features_87 + ) = ( + new_features_88 + ) = ( + new_features_89 + ) = ( + new_features_90 + ) = ( + new_features_91 + ) = ( + new_features_92 + ) = ( + new_features_93 + ) = new_features_94 = new_features_95 = new_features_96 = new_features_97 = None + x_400 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_401 = torch.nn.functional.relu(x_400, inplace=True) + x_400 = None + x_402 = torch.nn.functional.adaptive_avg_pool2d(x_401, 1) + x_401 = None + x_403 = x_402.flatten(1, -1) + x_402 = None + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = torch._C._nn.linear( + x_404, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_404 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_405,) diff --git a/samples/timm/densenet201.tv_in1k/weight_meta.py b/samples/timm/densenet201.tv_in1k/weight_meta.py new file mode 100644 index 000000000..d33ca6c09 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/weight_meta.py @@ -0,0 +1,10090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_" + shape = [128, 1664, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_" + shape = [128, 1696, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_" + shape = [128, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_" + shape = [128, 1760, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [896, 1792, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1664, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1696, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1760, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1792, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1856, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1888, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1920] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt b/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt new file mode 100644 index 000000000..f267737c7 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +c19519861ce0a98446d72201993ea0c8e7662560a288da6ea9997a3e4ddae2c2 \ No newline at end of file diff --git a/samples/timm/densenetblur121d.ra_in1k/graph_net.json b/samples/timm/densenetblur121d.ra_in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenetblur121d.ra_in1k/input_meta.py b/samples/timm/densenetblur121d.ra_in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenetblur121d.ra_in1k/input_tensor_constraints.py b/samples/timm/densenetblur121d.ra_in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/densenetblur121d.ra_in1k/model.py b/samples/timm/densenetblur121d.ra_in1k/model.py new file mode 100644 index 000000000..087a06325 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/model.py @@ -0,0 +1,5090 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_pool0_modules_1_buffers_filt_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_conv1_parameters_weight_ = ( + L_self_modules_features_modules_conv1_parameters_weight_ + ) + l_self_modules_features_modules_norm1_buffers_running_mean_ = ( + L_self_modules_features_modules_norm1_buffers_running_mean_ + ) + l_self_modules_features_modules_norm1_buffers_running_var_ = ( + L_self_modules_features_modules_norm1_buffers_running_var_ + ) + l_self_modules_features_modules_norm1_parameters_weight_ = ( + L_self_modules_features_modules_norm1_parameters_weight_ + ) + l_self_modules_features_modules_norm1_parameters_bias_ = ( + L_self_modules_features_modules_norm1_parameters_bias_ + ) + l_self_modules_features_modules_conv2_parameters_weight_ = ( + L_self_modules_features_modules_conv2_parameters_weight_ + ) + l_self_modules_features_modules_norm2_buffers_running_mean_ = ( + L_self_modules_features_modules_norm2_buffers_running_mean_ + ) + l_self_modules_features_modules_norm2_buffers_running_var_ = ( + L_self_modules_features_modules_norm2_buffers_running_var_ + ) + l_self_modules_features_modules_norm2_parameters_weight_ = ( + L_self_modules_features_modules_norm2_parameters_weight_ + ) + l_self_modules_features_modules_norm2_parameters_bias_ = ( + L_self_modules_features_modules_norm2_parameters_bias_ + ) + l_self_modules_features_modules_pool0_modules_1_buffers_filt_ = ( + L_self_modules_features_modules_pool0_modules_1_buffers_filt_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.conv2d( + x_1, + l_self_modules_features_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_1 = l_self_modules_features_modules_conv1_parameters_weight_ = None + x_2 = torch.nn.functional.batch_norm( + input_2, + l_self_modules_features_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_norm1_parameters_weight_, + l_self_modules_features_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_2 = ( + l_self_modules_features_modules_norm1_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm1_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm1_parameters_weight_ + ) = l_self_modules_features_modules_norm1_parameters_bias_ = None + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + input_3 = torch.conv2d( + x_3, + l_self_modules_features_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_conv2_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_norm2_parameters_weight_, + l_self_modules_features_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = ( + l_self_modules_features_modules_norm2_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm2_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm2_parameters_weight_ + ) = l_self_modules_features_modules_norm2_parameters_bias_ = None + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + input_4 = torch.nn.functional.max_pool2d( + x_5, 3, 1, 1, 1, ceil_mode=False, return_indices=False + ) + x_5 = None + x_6 = torch._C._nn.pad(input_4, [1, 1, 1, 1], "reflect", None) + input_4 = None + input_5 = torch.conv2d( + x_6, + l_self_modules_features_modules_pool0_modules_1_buffers_filt_, + stride=2, + groups=64, + ) + x_6 = l_self_modules_features_modules_pool0_modules_1_buffers_filt_ = None + concated_features = torch.cat([input_5], 1) + x_7 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_8 = torch.nn.functional.relu(x_7, inplace=True) + x_7 = None + bottleneck_output = torch.conv2d( + x_8, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_10 = torch.nn.functional.relu(x_9, inplace=True) + x_9 = None + new_features = torch.conv2d( + x_10, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_10 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_5, new_features], 1) + x_11 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_12 = torch.nn.functional.relu(x_11, inplace=True) + x_11 = None + bottleneck_output_1 = torch.conv2d( + x_12, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_12 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_14 = torch.nn.functional.relu(x_13, inplace=True) + x_13 = None + new_features_1 = torch.conv2d( + x_14, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_5, new_features, new_features_1], 1) + x_15 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_16 = torch.nn.functional.relu(x_15, inplace=True) + x_15 = None + bottleneck_output_2 = torch.conv2d( + x_16, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_16 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_18 = torch.nn.functional.relu(x_17, inplace=True) + x_17 = None + new_features_2 = torch.conv2d( + x_18, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_5, new_features, new_features_1, new_features_2], 1 + ) + x_19 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_20 = torch.nn.functional.relu(x_19, inplace=True) + x_19 = None + bottleneck_output_3 = torch.conv2d( + x_20, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_21 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_22 = torch.nn.functional.relu(x_21, inplace=True) + x_21 = None + new_features_3 = torch.conv2d( + x_22, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_22 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_5, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_23 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_24 = torch.nn.functional.relu(x_23, inplace=True) + x_23 = None + bottleneck_output_4 = torch.conv2d( + x_24, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_24 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_26 = torch.nn.functional.relu(x_25, inplace=True) + x_25 = None + new_features_4 = torch.conv2d( + x_26, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_26 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_5, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_27 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_28 = torch.nn.functional.relu(x_27, inplace=True) + x_27 = None + bottleneck_output_5 = torch.conv2d( + x_28, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_30 = torch.nn.functional.relu(x_29, inplace=True) + x_29 = None + new_features_5 = torch.conv2d( + x_30, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_30 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_5 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_31 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_32 = torch.nn.functional.relu(x_31, inplace=True) + x_31 = None + input_7 = torch.conv2d( + x_32, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_32 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_6 = torch.cat([input_8], 1) + x_33 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_34 = torch.nn.functional.relu(x_33, inplace=True) + x_33 = None + bottleneck_output_6 = torch.conv2d( + x_34, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_35 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_36 = torch.nn.functional.relu(x_35, inplace=True) + x_35 = None + new_features_6 = torch.conv2d( + x_36, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_36 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_8, new_features_6], 1) + x_37 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_38 = torch.nn.functional.relu(x_37, inplace=True) + x_37 = None + bottleneck_output_7 = torch.conv2d( + x_38, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_40 = torch.nn.functional.relu(x_39, inplace=True) + x_39 = None + new_features_7 = torch.conv2d( + x_40, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_40 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_8, new_features_6, new_features_7], 1) + x_41 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_42 = torch.nn.functional.relu(x_41, inplace=True) + x_41 = None + bottleneck_output_8 = torch.conv2d( + x_42, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_44 = torch.nn.functional.relu(x_43, inplace=True) + x_43 = None + new_features_8 = torch.conv2d( + x_44, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_8, new_features_6, new_features_7, new_features_8], 1 + ) + x_45 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_46 = torch.nn.functional.relu(x_45, inplace=True) + x_45 = None + bottleneck_output_9 = torch.conv2d( + x_46, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_47 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_48 = torch.nn.functional.relu(x_47, inplace=True) + x_47 = None + new_features_9 = torch.conv2d( + x_48, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_8, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_49 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_50 = torch.nn.functional.relu(x_49, inplace=True) + x_49 = None + bottleneck_output_10 = torch.conv2d( + x_50, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_50 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_52 = torch.nn.functional.relu(x_51, inplace=True) + x_51 = None + new_features_10 = torch.conv2d( + x_52, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_52 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_53 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_54 = torch.nn.functional.relu(x_53, inplace=True) + x_53 = None + bottleneck_output_11 = torch.conv2d( + x_54, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_54 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_55 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_56 = torch.nn.functional.relu(x_55, inplace=True) + x_55 = None + new_features_11 = torch.conv2d( + x_56, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_57 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_58 = torch.nn.functional.relu(x_57, inplace=True) + x_57 = None + bottleneck_output_12 = torch.conv2d( + x_58, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_58 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_59 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_60 = torch.nn.functional.relu(x_59, inplace=True) + x_59 = None + new_features_12 = torch.conv2d( + x_60, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_60 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_61 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_62 = torch.nn.functional.relu(x_61, inplace=True) + x_61 = None + bottleneck_output_13 = torch.conv2d( + x_62, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_62 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_64 = torch.nn.functional.relu(x_63, inplace=True) + x_63 = None + new_features_13 = torch.conv2d( + x_64, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_65 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_66 = torch.nn.functional.relu(x_65, inplace=True) + x_65 = None + bottleneck_output_14 = torch.conv2d( + x_66, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_66 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_68 = torch.nn.functional.relu(x_67, inplace=True) + x_67 = None + new_features_14 = torch.conv2d( + x_68, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_68 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_69 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_70 = torch.nn.functional.relu(x_69, inplace=True) + x_69 = None + bottleneck_output_15 = torch.conv2d( + x_70, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_71 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_72 = torch.nn.functional.relu(x_71, inplace=True) + x_71 = None + new_features_15 = torch.conv2d( + x_72, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_73 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_74 = torch.nn.functional.relu(x_73, inplace=True) + x_73 = None + bottleneck_output_16 = torch.conv2d( + x_74, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_75 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_76 = torch.nn.functional.relu(x_75, inplace=True) + x_75 = None + new_features_16 = torch.conv2d( + x_76, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_76 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_77 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_78 = torch.nn.functional.relu(x_77, inplace=True) + x_77 = None + bottleneck_output_17 = torch.conv2d( + x_78, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_80 = torch.nn.functional.relu(x_79, inplace=True) + x_79 = None + new_features_17 = torch.conv2d( + x_80, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_8 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_81 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_82 = torch.nn.functional.relu(x_81, inplace=True) + x_81 = None + input_10 = torch.conv2d( + x_82, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_18 = torch.cat([input_11], 1) + x_83 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_84 = torch.nn.functional.relu(x_83, inplace=True) + x_83 = None + bottleneck_output_18 = torch.conv2d( + x_84, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_86 = torch.nn.functional.relu(x_85, inplace=True) + x_85 = None + new_features_18 = torch.conv2d( + x_86, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_11, new_features_18], 1) + x_87 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_88 = torch.nn.functional.relu(x_87, inplace=True) + x_87 = None + bottleneck_output_19 = torch.conv2d( + x_88, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_89 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_90 = torch.nn.functional.relu(x_89, inplace=True) + x_89 = None + new_features_19 = torch.conv2d( + x_90, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_90 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat( + [input_11, new_features_18, new_features_19], 1 + ) + x_91 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_92 = torch.nn.functional.relu(x_91, inplace=True) + x_91 = None + bottleneck_output_20 = torch.conv2d( + x_92, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_93 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_94 = torch.nn.functional.relu(x_93, inplace=True) + x_93 = None + new_features_20 = torch.conv2d( + x_94, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_11, new_features_18, new_features_19, new_features_20], 1 + ) + x_95 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_96 = torch.nn.functional.relu(x_95, inplace=True) + x_95 = None + bottleneck_output_21 = torch.conv2d( + x_96, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_98 = torch.nn.functional.relu(x_97, inplace=True) + x_97 = None + new_features_21 = torch.conv2d( + x_98, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_98 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_99 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_100 = torch.nn.functional.relu(x_99, inplace=True) + x_99 = None + bottleneck_output_22 = torch.conv2d( + x_100, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_102 = torch.nn.functional.relu(x_101, inplace=True) + x_101 = None + new_features_22 = torch.conv2d( + x_102, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_103 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_104 = torch.nn.functional.relu(x_103, inplace=True) + x_103 = None + bottleneck_output_23 = torch.conv2d( + x_104, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_106 = torch.nn.functional.relu(x_105, inplace=True) + x_105 = None + new_features_23 = torch.conv2d( + x_106, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_107 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_108 = torch.nn.functional.relu(x_107, inplace=True) + x_107 = None + bottleneck_output_24 = torch.conv2d( + x_108, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_108 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_109 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_110 = torch.nn.functional.relu(x_109, inplace=True) + x_109 = None + new_features_24 = torch.conv2d( + x_110, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_110 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_111 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_112 = torch.nn.functional.relu(x_111, inplace=True) + x_111 = None + bottleneck_output_25 = torch.conv2d( + x_112, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_113 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_114 = torch.nn.functional.relu(x_113, inplace=True) + x_113 = None + new_features_25 = torch.conv2d( + x_114, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_114 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_115 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_116 = torch.nn.functional.relu(x_115, inplace=True) + x_115 = None + bottleneck_output_26 = torch.conv2d( + x_116, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_118 = torch.nn.functional.relu(x_117, inplace=True) + x_117 = None + new_features_26 = torch.conv2d( + x_118, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_118 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_119 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_120 = torch.nn.functional.relu(x_119, inplace=True) + x_119 = None + bottleneck_output_27 = torch.conv2d( + x_120, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_122 = torch.nn.functional.relu(x_121, inplace=True) + x_121 = None + new_features_27 = torch.conv2d( + x_122, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_123 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_124 = torch.nn.functional.relu(x_123, inplace=True) + x_123 = None + bottleneck_output_28 = torch.conv2d( + x_124, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_124 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_126 = torch.nn.functional.relu(x_125, inplace=True) + x_125 = None + new_features_28 = torch.conv2d( + x_126, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_126 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_127 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_128 = torch.nn.functional.relu(x_127, inplace=True) + x_127 = None + bottleneck_output_29 = torch.conv2d( + x_128, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_129 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_130 = torch.nn.functional.relu(x_129, inplace=True) + x_129 = None + new_features_29 = torch.conv2d( + x_130, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_131 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_132 = torch.nn.functional.relu(x_131, inplace=True) + x_131 = None + bottleneck_output_30 = torch.conv2d( + x_132, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_133 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_134 = torch.nn.functional.relu(x_133, inplace=True) + x_133 = None + new_features_30 = torch.conv2d( + x_134, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_134 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_135 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_136 = torch.nn.functional.relu(x_135, inplace=True) + x_135 = None + bottleneck_output_31 = torch.conv2d( + x_136, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_137 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_138 = torch.nn.functional.relu(x_137, inplace=True) + x_137 = None + new_features_31 = torch.conv2d( + x_138, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_138 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_139 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_140 = torch.nn.functional.relu(x_139, inplace=True) + x_139 = None + bottleneck_output_32 = torch.conv2d( + x_140, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_142 = torch.nn.functional.relu(x_141, inplace=True) + x_141 = None + new_features_32 = torch.conv2d( + x_142, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_142 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_143 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_144 = torch.nn.functional.relu(x_143, inplace=True) + x_143 = None + bottleneck_output_33 = torch.conv2d( + x_144, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_146 = torch.nn.functional.relu(x_145, inplace=True) + x_145 = None + new_features_33 = torch.conv2d( + x_146, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_146 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_147 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_148 = torch.nn.functional.relu(x_147, inplace=True) + x_147 = None + bottleneck_output_34 = torch.conv2d( + x_148, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_148 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_150 = torch.nn.functional.relu(x_149, inplace=True) + x_149 = None + new_features_34 = torch.conv2d( + x_150, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_150 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_151 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_152 = torch.nn.functional.relu(x_151, inplace=True) + x_151 = None + bottleneck_output_35 = torch.conv2d( + x_152, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_152 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_154 = torch.nn.functional.relu(x_153, inplace=True) + x_153 = None + new_features_35 = torch.conv2d( + x_154, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_154 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_155 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_156 = torch.nn.functional.relu(x_155, inplace=True) + x_155 = None + bottleneck_output_36 = torch.conv2d( + x_156, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_156 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_158 = torch.nn.functional.relu(x_157, inplace=True) + x_157 = None + new_features_36 = torch.conv2d( + x_158, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_159 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_160 = torch.nn.functional.relu(x_159, inplace=True) + x_159 = None + bottleneck_output_37 = torch.conv2d( + x_160, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_160 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_161 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_162 = torch.nn.functional.relu(x_161, inplace=True) + x_161 = None + new_features_37 = torch.conv2d( + x_162, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_162 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_163 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_164 = torch.nn.functional.relu(x_163, inplace=True) + x_163 = None + bottleneck_output_38 = torch.conv2d( + x_164, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_164 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_166 = torch.nn.functional.relu(x_165, inplace=True) + x_165 = None + new_features_38 = torch.conv2d( + x_166, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_166 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_167 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_168 = torch.nn.functional.relu(x_167, inplace=True) + x_167 = None + bottleneck_output_39 = torch.conv2d( + x_168, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_168 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_170 = torch.nn.functional.relu(x_169, inplace=True) + x_169 = None + new_features_39 = torch.conv2d( + x_170, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_170 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_171 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_172 = torch.nn.functional.relu(x_171, inplace=True) + x_171 = None + bottleneck_output_40 = torch.conv2d( + x_172, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_172 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_173 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_174 = torch.nn.functional.relu(x_173, inplace=True) + x_173 = None + new_features_40 = torch.conv2d( + x_174, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_175 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_176 = torch.nn.functional.relu(x_175, inplace=True) + x_175 = None + bottleneck_output_41 = torch.conv2d( + x_176, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_177 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_178 = torch.nn.functional.relu(x_177, inplace=True) + x_177 = None + new_features_41 = torch.conv2d( + x_178, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_178 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + input_11 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = new_features_38 = new_features_39 = new_features_40 = new_features_41 = None + x_179 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_180 = torch.nn.functional.relu(x_179, inplace=True) + x_179 = None + input_13 = torch.conv2d( + x_180, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_14 = torch._C._nn.avg_pool2d(input_13, 2, 2, 0, False, True, None) + input_13 = None + concated_features_42 = torch.cat([input_14], 1) + x_181 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_182 = torch.nn.functional.relu(x_181, inplace=True) + x_181 = None + bottleneck_output_42 = torch.conv2d( + x_182, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_183 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_184 = torch.nn.functional.relu(x_183, inplace=True) + x_183 = None + new_features_42 = torch.conv2d( + x_184, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_184 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat([input_14, new_features_42], 1) + x_185 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_186 = torch.nn.functional.relu(x_185, inplace=True) + x_185 = None + bottleneck_output_43 = torch.conv2d( + x_186, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_186 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_188 = torch.nn.functional.relu(x_187, inplace=True) + x_187 = None + new_features_43 = torch.conv2d( + x_188, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_188 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [input_14, new_features_42, new_features_43], 1 + ) + x_189 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_190 = torch.nn.functional.relu(x_189, inplace=True) + x_189 = None + bottleneck_output_44 = torch.conv2d( + x_190, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_190 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_191 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_192 = torch.nn.functional.relu(x_191, inplace=True) + x_191 = None + new_features_44 = torch.conv2d( + x_192, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_192 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [input_14, new_features_42, new_features_43, new_features_44], 1 + ) + x_193 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_194 = torch.nn.functional.relu(x_193, inplace=True) + x_193 = None + bottleneck_output_45 = torch.conv2d( + x_194, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_194 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_196 = torch.nn.functional.relu(x_195, inplace=True) + x_195 = None + new_features_45 = torch.conv2d( + x_196, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_196 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_197 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_198 = torch.nn.functional.relu(x_197, inplace=True) + x_197 = None + bottleneck_output_46 = torch.conv2d( + x_198, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_198 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_200 = torch.nn.functional.relu(x_199, inplace=True) + x_199 = None + new_features_46 = torch.conv2d( + x_200, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_201 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_202 = torch.nn.functional.relu(x_201, inplace=True) + x_201 = None + bottleneck_output_47 = torch.conv2d( + x_202, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_203 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_204 = torch.nn.functional.relu(x_203, inplace=True) + x_203 = None + new_features_47 = torch.conv2d( + x_204, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_204 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_205 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_206 = torch.nn.functional.relu(x_205, inplace=True) + x_205 = None + bottleneck_output_48 = torch.conv2d( + x_206, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_206 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_207 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_208 = torch.nn.functional.relu(x_207, inplace=True) + x_207 = None + new_features_48 = torch.conv2d( + x_208, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_208 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_209 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_210 = torch.nn.functional.relu(x_209, inplace=True) + x_209 = None + bottleneck_output_49 = torch.conv2d( + x_210, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_210 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_211 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_212 = torch.nn.functional.relu(x_211, inplace=True) + x_211 = None + new_features_49 = torch.conv2d( + x_212, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_212 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_213 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_214 = torch.nn.functional.relu(x_213, inplace=True) + x_213 = None + bottleneck_output_50 = torch.conv2d( + x_214, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_215 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_216 = torch.nn.functional.relu(x_215, inplace=True) + x_215 = None + new_features_50 = torch.conv2d( + x_216, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_216 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_217 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_218 = torch.nn.functional.relu(x_217, inplace=True) + x_217 = None + bottleneck_output_51 = torch.conv2d( + x_218, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_218 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_219 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_220 = torch.nn.functional.relu(x_219, inplace=True) + x_219 = None + new_features_51 = torch.conv2d( + x_220, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_220 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_221 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_222 = torch.nn.functional.relu(x_221, inplace=True) + x_221 = None + bottleneck_output_52 = torch.conv2d( + x_222, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_223 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_224 = torch.nn.functional.relu(x_223, inplace=True) + x_223 = None + new_features_52 = torch.conv2d( + x_224, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_224 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_225 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_226 = torch.nn.functional.relu(x_225, inplace=True) + x_225 = None + bottleneck_output_53 = torch.conv2d( + x_226, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_226 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_227 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_228 = torch.nn.functional.relu(x_227, inplace=True) + x_227 = None + new_features_53 = torch.conv2d( + x_228, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_228 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_229 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_230 = torch.nn.functional.relu(x_229, inplace=True) + x_229 = None + bottleneck_output_54 = torch.conv2d( + x_230, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_230 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_231 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_232 = torch.nn.functional.relu(x_231, inplace=True) + x_231 = None + new_features_54 = torch.conv2d( + x_232, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_232 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_233 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_234 = torch.nn.functional.relu(x_233, inplace=True) + x_233 = None + bottleneck_output_55 = torch.conv2d( + x_234, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_235 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_236 = torch.nn.functional.relu(x_235, inplace=True) + x_235 = None + new_features_55 = torch.conv2d( + x_236, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_236 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_237 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_238 = torch.nn.functional.relu(x_237, inplace=True) + x_237 = None + bottleneck_output_56 = torch.conv2d( + x_238, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_238 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_239 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_240 = torch.nn.functional.relu(x_239, inplace=True) + x_239 = None + new_features_56 = torch.conv2d( + x_240, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_240 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_241 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_242 = torch.nn.functional.relu(x_241, inplace=True) + x_241 = None + bottleneck_output_57 = torch.conv2d( + x_242, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_243 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_244 = torch.nn.functional.relu(x_243, inplace=True) + x_243 = None + new_features_57 = torch.conv2d( + x_244, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_244 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + input_15 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + input_14 = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = new_features_54 = new_features_55 = new_features_56 = new_features_57 = None + x_245 = torch.nn.functional.batch_norm( + input_15, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_15 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_246 = torch.nn.functional.relu(x_245, inplace=True) + x_245 = None + x_247 = torch.nn.functional.adaptive_avg_pool2d(x_246, 1) + x_246 = None + x_248 = x_247.flatten(1, -1) + x_247 = None + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_249 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_250,) diff --git a/samples/timm/densenetblur121d.ra_in1k/weight_meta.py b/samples/timm/densenetblur121d.ra_in1k/weight_meta.py new file mode 100644 index 000000000..487168616 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/weight_meta.py @@ -0,0 +1,6200 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.282 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_conv1_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_norm1_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_norm1_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_norm1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_norm1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_conv2_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_norm2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_norm2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_pool0_modules_1_buffers_filt_: + name = "L_self_modules_features_modules_pool0_modules_1_buffers_filt_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.111 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/dla34.in1k/graph_hash.txt b/samples/timm/dla34.in1k/graph_hash.txt new file mode 100644 index 000000000..793809316 --- /dev/null +++ b/samples/timm/dla34.in1k/graph_hash.txt @@ -0,0 +1 @@ +13e26b9a256b6c4b6ff13fdc717e0de35f09711d723103383755a556a61ce457 \ No newline at end of file diff --git a/samples/timm/dla34.in1k/graph_net.json b/samples/timm/dla34.in1k/graph_net.json new file mode 100644 index 000000000..1373fe3b5 --- /dev/null +++ b/samples/timm/dla34.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/dla34.in1k/input_meta.py b/samples/timm/dla34.in1k/input_meta.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/dla34.in1k/input_tensor_constraints.py b/samples/timm/dla34.in1k/input_tensor_constraints.py new file mode 100644 index 000000000..e69de29bb diff --git a/samples/timm/dla34.in1k/model.py b/samples/timm/dla34.in1k/model.py new file mode 100644 index 000000000..21ca6551e --- /dev/null +++ b/samples/timm/dla34.in1k/model.py @@ -0,0 +1,1596 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_base_layer_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_base_layer_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_base_layer_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_base_layer_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_base_layer_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level0_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level0_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level1_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level1_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_base_layer_modules_0_parameters_weight_ = ( + L_self_modules_base_layer_modules_0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_base_layer_modules_1_buffers_running_mean_ = ( + L_self_modules_base_layer_modules_1_buffers_running_mean_ + ) + l_self_modules_base_layer_modules_1_buffers_running_var_ = ( + L_self_modules_base_layer_modules_1_buffers_running_var_ + ) + l_self_modules_base_layer_modules_1_parameters_weight_ = ( + L_self_modules_base_layer_modules_1_parameters_weight_ + ) + l_self_modules_base_layer_modules_1_parameters_bias_ = ( + L_self_modules_base_layer_modules_1_parameters_bias_ + ) + l_self_modules_level0_modules_0_parameters_weight_ = ( + L_self_modules_level0_modules_0_parameters_weight_ + ) + l_self_modules_level0_modules_1_buffers_running_mean_ = ( + L_self_modules_level0_modules_1_buffers_running_mean_ + ) + l_self_modules_level0_modules_1_buffers_running_var_ = ( + L_self_modules_level0_modules_1_buffers_running_var_ + ) + l_self_modules_level0_modules_1_parameters_weight_ = ( + L_self_modules_level0_modules_1_parameters_weight_ + ) + l_self_modules_level0_modules_1_parameters_bias_ = ( + L_self_modules_level0_modules_1_parameters_bias_ + ) + l_self_modules_level1_modules_0_parameters_weight_ = ( + L_self_modules_level1_modules_0_parameters_weight_ + ) + l_self_modules_level1_modules_1_buffers_running_mean_ = ( + L_self_modules_level1_modules_1_buffers_running_mean_ + ) + l_self_modules_level1_modules_1_buffers_running_var_ = ( + L_self_modules_level1_modules_1_buffers_running_var_ + ) + l_self_modules_level1_modules_1_parameters_weight_ = ( + L_self_modules_level1_modules_1_parameters_weight_ + ) + l_self_modules_level1_modules_1_parameters_bias_ = ( + L_self_modules_level1_modules_1_parameters_bias_ + ) + l_self_modules_level2_modules_project_modules_0_parameters_weight_ = ( + L_self_modules_level2_modules_project_modules_0_parameters_weight_ + ) + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_ = ( + L_self_modules_level2_modules_project_modules_1_buffers_running_mean_ + ) + l_self_modules_level2_modules_project_modules_1_buffers_running_var_ = ( + L_self_modules_level2_modules_project_modules_1_buffers_running_var_ + ) + l_self_modules_level2_modules_project_modules_1_parameters_weight_ = ( + L_self_modules_level2_modules_project_modules_1_parameters_weight_ + ) + l_self_modules_level2_modules_project_modules_1_parameters_bias_ = ( + L_self_modules_level2_modules_project_modules_1_parameters_bias_ + ) + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ + ) + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ + ) + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ + ) + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ + ) + l_self_modules_level2_modules_root_modules_conv_parameters_weight_ = ( + L_self_modules_level2_modules_root_modules_conv_parameters_weight_ + ) + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ = ( + L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ + ) + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_ = ( + L_self_modules_level2_modules_root_modules_bn_buffers_running_var_ + ) + l_self_modules_level2_modules_root_modules_bn_parameters_weight_ = ( + L_self_modules_level2_modules_root_modules_bn_parameters_weight_ + ) + l_self_modules_level2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level5_modules_project_modules_0_parameters_weight_ = ( + L_self_modules_level5_modules_project_modules_0_parameters_weight_ + ) + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_ = ( + L_self_modules_level5_modules_project_modules_1_buffers_running_mean_ + ) + l_self_modules_level5_modules_project_modules_1_buffers_running_var_ = ( + L_self_modules_level5_modules_project_modules_1_buffers_running_var_ + ) + l_self_modules_level5_modules_project_modules_1_parameters_weight_ = ( + L_self_modules_level5_modules_project_modules_1_parameters_weight_ + ) + l_self_modules_level5_modules_project_modules_1_parameters_bias_ = ( + L_self_modules_level5_modules_project_modules_1_parameters_bias_ + ) + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ + ) + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ + ) + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ + ) + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ + ) + l_self_modules_level5_modules_root_modules_conv_parameters_weight_ = ( + L_self_modules_level5_modules_root_modules_conv_parameters_weight_ + ) + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ = ( + L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ + ) + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_ = ( + L_self_modules_level5_modules_root_modules_bn_buffers_running_var_ + ) + l_self_modules_level5_modules_root_modules_bn_parameters_weight_ = ( + L_self_modules_level5_modules_root_modules_bn_parameters_weight_ + ) + l_self_modules_level5_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level5_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_fc_parameters_weight_ = L_self_modules_fc_parameters_weight_ + l_self_modules_fc_parameters_bias_ = L_self_modules_fc_parameters_bias_ + input_1 = torch.conv2d( + l_x_, + l_self_modules_base_layer_modules_0_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_base_layer_modules_0_parameters_weight_ = None + input_2 = torch.nn.functional.batch_norm( + input_1, + l_self_modules_base_layer_modules_1_buffers_running_mean_, + l_self_modules_base_layer_modules_1_buffers_running_var_, + l_self_modules_base_layer_modules_1_parameters_weight_, + l_self_modules_base_layer_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_base_layer_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_base_layer_modules_1_buffers_running_var_ + ) = ( + l_self_modules_base_layer_modules_1_parameters_weight_ + ) = l_self_modules_base_layer_modules_1_parameters_bias_ = None + input_3 = torch.nn.functional.relu(input_2, inplace=True) + input_2 = None + input_4 = torch.conv2d( + input_3, + l_self_modules_level0_modules_0_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_3 = l_self_modules_level0_modules_0_parameters_weight_ = None + input_5 = torch.nn.functional.batch_norm( + input_4, + l_self_modules_level0_modules_1_buffers_running_mean_, + l_self_modules_level0_modules_1_buffers_running_var_, + l_self_modules_level0_modules_1_parameters_weight_, + l_self_modules_level0_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_4 = ( + l_self_modules_level0_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level0_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level0_modules_1_parameters_weight_ + ) = l_self_modules_level0_modules_1_parameters_bias_ = None + input_6 = torch.nn.functional.relu(input_5, inplace=True) + input_5 = None + input_7 = torch.conv2d( + input_6, + l_self_modules_level1_modules_0_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_6 = l_self_modules_level1_modules_0_parameters_weight_ = None + input_8 = torch.nn.functional.batch_norm( + input_7, + l_self_modules_level1_modules_1_buffers_running_mean_, + l_self_modules_level1_modules_1_buffers_running_var_, + l_self_modules_level1_modules_1_parameters_weight_, + l_self_modules_level1_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_7 = ( + l_self_modules_level1_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level1_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level1_modules_1_parameters_weight_ + ) = l_self_modules_level1_modules_1_parameters_bias_ = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + bottom = torch.nn.functional.max_pool2d( + input_9, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_10 = torch.conv2d( + bottom, + l_self_modules_level2_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom = ( + l_self_modules_level2_modules_project_modules_0_parameters_weight_ + ) = None + input_11 = torch.nn.functional.batch_norm( + input_10, + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level2_modules_project_modules_1_buffers_running_var_, + l_self_modules_level2_modules_project_modules_1_parameters_weight_, + l_self_modules_level2_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_10 = ( + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_project_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_project_modules_1_parameters_weight_ + ) = l_self_modules_level2_modules_project_modules_1_parameters_bias_ = None + out = torch.conv2d( + input_9, + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = ( + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ + ) = None + out_1 = torch.nn.functional.batch_norm( + out, + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out = ( + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ + ) = l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ = None + out_2 = torch.nn.functional.relu(out_1, inplace=True) + out_1 = None + out_3 = torch.conv2d( + out_2, + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_2 = ( + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ + ) = None + out_4 = torch.nn.functional.batch_norm( + out_3, + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_3 = ( + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ + ) = l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ = None + out_4 += input_11 + out_5 = out_4 + out_4 = input_11 = None + out_6 = torch.nn.functional.relu(out_5, inplace=True) + out_5 = None + out_7 = torch.conv2d( + out_6, + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ = None + out_8 = torch.nn.functional.batch_norm( + out_7, + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_7 = ( + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ + ) = l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ = None + out_9 = torch.nn.functional.relu(out_8, inplace=True) + out_8 = None + out_10 = torch.conv2d( + out_9, + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_9 = ( + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ + ) = None + out_11 = torch.nn.functional.batch_norm( + out_10, + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_10 = ( + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ + ) = l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ = None + out_11 += out_6 + out_12 = out_11 + out_11 = None + out_13 = torch.nn.functional.relu(out_12, inplace=True) + out_12 = None + cat = torch.cat([out_13, out_6], 1) + out_13 = out_6 = None + x = torch.conv2d( + cat, + l_self_modules_level2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_level2_modules_root_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_root_modules_bn_parameters_weight_ + ) = l_self_modules_level2_modules_root_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.relu(x_1, inplace=True) + x_1 = None + bottom_1 = torch.nn.functional.max_pool2d( + x_2, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + bottom_2 = torch.nn.functional.max_pool2d( + x_2, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_12 = torch.conv2d( + bottom_2, + l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom_2 = l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ = (None) + input_13 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ = (None) + out_14 = torch.conv2d( + x_2, + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = (None) + out_15 = torch.nn.functional.batch_norm( + out_14, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_14 = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_16 = torch.nn.functional.relu(out_15, inplace=True) + out_15 = None + out_17 = torch.conv2d( + out_16, + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_16 = l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_18 = torch.nn.functional.batch_norm( + out_17, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_17 = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_18 += input_13 + out_19 = out_18 + out_18 = input_13 = None + out_20 = torch.nn.functional.relu(out_19, inplace=True) + out_19 = None + out_21 = torch.conv2d( + out_20, + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_22 = torch.nn.functional.batch_norm( + out_21, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_21 = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_23 = torch.nn.functional.relu(out_22, inplace=True) + out_22 = None + out_24 = torch.conv2d( + out_23, + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_23 = l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_25 = torch.nn.functional.batch_norm( + out_24, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_24 = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_25 += out_20 + out_26 = out_25 + out_25 = None + out_27 = torch.nn.functional.relu(out_26, inplace=True) + out_26 = None + cat_1 = torch.cat([out_27, out_20], 1) + out_27 = out_20 = None + x_3 = torch.conv2d( + cat_1, + l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) = None + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + out_28 = torch.conv2d( + x_5, + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = ( + None + ) + out_29 = torch.nn.functional.batch_norm( + out_28, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_28 = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_30 = torch.nn.functional.relu(out_29, inplace=True) + out_29 = None + out_31 = torch.conv2d( + out_30, + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_30 = l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_32 = torch.nn.functional.batch_norm( + out_31, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_31 = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_32 += x_5 + out_33 = out_32 + out_32 = None + out_34 = torch.nn.functional.relu(out_33, inplace=True) + out_33 = None + out_35 = torch.conv2d( + out_34, + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_36 = torch.nn.functional.batch_norm( + out_35, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_35 = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_37 = torch.nn.functional.relu(out_36, inplace=True) + out_36 = None + out_38 = torch.conv2d( + out_37, + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_37 = l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_39 = torch.nn.functional.batch_norm( + out_38, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_38 = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_39 += out_34 + out_40 = out_39 + out_39 = None + out_41 = torch.nn.functional.relu(out_40, inplace=True) + out_40 = None + cat_2 = torch.cat([out_41, out_34, bottom_1, x_5], 1) + out_41 = out_34 = bottom_1 = x_5 = None + x_6 = torch.conv2d( + cat_2, + l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.relu(x_7, inplace=True) + x_7 = None + bottom_3 = torch.nn.functional.max_pool2d( + x_8, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + bottom_4 = torch.nn.functional.max_pool2d( + x_8, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_14 = torch.conv2d( + bottom_4, + l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom_4 = l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ = (None) + input_15 = torch.nn.functional.batch_norm( + input_14, + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_14 = l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ = (None) + out_42 = torch.conv2d( + x_8, + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = (None) + out_43 = torch.nn.functional.batch_norm( + out_42, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_42 = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_44 = torch.nn.functional.relu(out_43, inplace=True) + out_43 = None + out_45 = torch.conv2d( + out_44, + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_44 = l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_46 = torch.nn.functional.batch_norm( + out_45, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_45 = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_46 += input_15 + out_47 = out_46 + out_46 = input_15 = None + out_48 = torch.nn.functional.relu(out_47, inplace=True) + out_47 = None + out_49 = torch.conv2d( + out_48, + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_50 = torch.nn.functional.batch_norm( + out_49, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_49 = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_51 = torch.nn.functional.relu(out_50, inplace=True) + out_50 = None + out_52 = torch.conv2d( + out_51, + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_51 = l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_53 = torch.nn.functional.batch_norm( + out_52, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_52 = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_53 += out_48 + out_54 = out_53 + out_53 = None + out_55 = torch.nn.functional.relu(out_54, inplace=True) + out_54 = None + cat_3 = torch.cat([out_55, out_48], 1) + out_55 = out_48 = None + x_9 = torch.conv2d( + cat_3, + l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + out_56 = torch.conv2d( + x_11, + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = ( + None + ) + out_57 = torch.nn.functional.batch_norm( + out_56, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_56 = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_58 = torch.nn.functional.relu(out_57, inplace=True) + out_57 = None + out_59 = torch.conv2d( + out_58, + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_58 = l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_60 = torch.nn.functional.batch_norm( + out_59, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_59 = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_60 += x_11 + out_61 = out_60 + out_60 = None + out_62 = torch.nn.functional.relu(out_61, inplace=True) + out_61 = None + out_63 = torch.conv2d( + out_62, + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_64 = torch.nn.functional.batch_norm( + out_63, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_63 = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_65 = torch.nn.functional.relu(out_64, inplace=True) + out_64 = None + out_66 = torch.conv2d( + out_65, + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_65 = l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_67 = torch.nn.functional.batch_norm( + out_66, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_66 = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_67 += out_62 + out_68 = out_67 + out_67 = None + out_69 = torch.nn.functional.relu(out_68, inplace=True) + out_68 = None + cat_4 = torch.cat([out_69, out_62, bottom_3, x_11], 1) + out_69 = out_62 = bottom_3 = x_11 = None + x_12 = torch.conv2d( + cat_4, + l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_4 = l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) = None + x_14 = torch.nn.functional.relu(x_13, inplace=True) + x_13 = None + bottom_5 = torch.nn.functional.max_pool2d( + x_14, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_16 = torch.conv2d( + bottom_5, + l_self_modules_level5_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_level5_modules_project_modules_0_parameters_weight_ = None + input_17 = torch.nn.functional.batch_norm( + input_16, + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level5_modules_project_modules_1_buffers_running_var_, + l_self_modules_level5_modules_project_modules_1_parameters_weight_, + l_self_modules_level5_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_16 = ( + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_project_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_project_modules_1_parameters_weight_ + ) = l_self_modules_level5_modules_project_modules_1_parameters_bias_ = None + out_70 = torch.conv2d( + x_14, + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_14 = ( + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ + ) = None + out_71 = torch.nn.functional.batch_norm( + out_70, + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_70 = ( + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ + ) = l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ = None + out_72 = torch.nn.functional.relu(out_71, inplace=True) + out_71 = None + out_73 = torch.conv2d( + out_72, + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_72 = ( + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ + ) = None + out_74 = torch.nn.functional.batch_norm( + out_73, + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_73 = ( + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ + ) = l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ = None + out_74 += input_17 + out_75 = out_74 + out_74 = input_17 = None + out_76 = torch.nn.functional.relu(out_75, inplace=True) + out_75 = None + out_77 = torch.conv2d( + out_76, + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ = None + out_78 = torch.nn.functional.batch_norm( + out_77, + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_77 = ( + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ + ) = l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ = None + out_79 = torch.nn.functional.relu(out_78, inplace=True) + out_78 = None + out_80 = torch.conv2d( + out_79, + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_79 = ( + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ + ) = None + out_81 = torch.nn.functional.batch_norm( + out_80, + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_80 = ( + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ + ) = l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ = None + out_81 += out_76 + out_82 = out_81 + out_81 = None + out_83 = torch.nn.functional.relu(out_82, inplace=True) + out_82 = None + cat_5 = torch.cat([out_83, out_76, bottom_5], 1) + out_83 = out_76 = bottom_5 = None + x_15 = torch.conv2d( + cat_5, + l_self_modules_level5_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_5 = ( + l_self_modules_level5_modules_root_modules_conv_parameters_weight_ + ) = None + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level5_modules_root_modules_bn_parameters_weight_, + l_self_modules_level5_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = ( + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_root_modules_bn_parameters_weight_ + ) = l_self_modules_level5_modules_root_modules_bn_parameters_bias_ = None + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + x_18 = torch.nn.functional.adaptive_avg_pool2d(x_17, 1) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_fc_parameters_weight_, + l_self_modules_fc_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = ( + l_self_modules_fc_parameters_weight_ + ) = l_self_modules_fc_parameters_bias_ = None + x_21 = x_20.flatten(1, -1) + x_20 = None + return (x_21,) diff --git a/samples/timm/dla34.in1k/weight_meta.py b/samples/timm/dla34.in1k/weight_meta.py new file mode 100644 index 000000000..5c0e8c330 --- /dev/null +++ b/samples/timm/dla34.in1k/weight_meta.py @@ -0,0 +1,1940 @@ +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_0_parameters_weight_: + name = "L_self_modules_base_layer_modules_0_parameters_weight_" + shape = [16, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_buffers_running_mean_: + name = "L_self_modules_base_layer_modules_1_buffers_running_mean_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_buffers_running_var_: + name = "L_self_modules_base_layer_modules_1_buffers_running_var_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_parameters_weight_: + name = "L_self_modules_base_layer_modules_1_parameters_weight_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_parameters_bias_: + name = "L_self_modules_base_layer_modules_1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_0_parameters_weight_: + name = "L_self_modules_level0_modules_0_parameters_weight_" + shape = [16, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_buffers_running_mean_: + name = "L_self_modules_level0_modules_1_buffers_running_mean_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_buffers_running_var_: + name = "L_self_modules_level0_modules_1_buffers_running_var_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_parameters_weight_: + name = "L_self_modules_level0_modules_1_parameters_weight_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_parameters_bias_: + name = "L_self_modules_level0_modules_1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_0_parameters_weight_: + name = "L_self_modules_level1_modules_0_parameters_weight_" + shape = [32, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_buffers_running_mean_: + name = "L_self_modules_level1_modules_1_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_buffers_running_var_: + name = "L_self_modules_level1_modules_1_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_parameters_weight_: + name = "L_self_modules_level1_modules_1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_parameters_bias_: + name = "L_self_modules_level1_modules_1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level2_modules_project_modules_0_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level2_modules_project_modules_1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level2_modules_project_modules_1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level2_modules_project_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_parameters_bias_: + name = "L_self_modules_level2_modules_project_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level2_modules_root_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level2_modules_root_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_parameters_weight_: + name = "L_self_modules_level2_modules_root_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_parameters_bias_: + name = "L_self_modules_level2_modules_root_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_" + shape = [256, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level5_modules_project_modules_0_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level5_modules_project_modules_1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level5_modules_project_modules_1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level5_modules_project_modules_1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_parameters_bias_: + name = "L_self_modules_level5_modules_project_modules_1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level5_modules_root_modules_conv_parameters_weight_" + shape = [512, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level5_modules_root_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_parameters_weight_: + name = "L_self_modules_level5_modules_root_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_parameters_bias_: + name = "L_self_modules_level5_modules_root_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_parameters_weight_: + name = "L_self_modules_fc_parameters_weight_" + shape = [1000, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_parameters_bias_: + name = "L_self_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.025 + data = None